Method, device and readable medium for calculating the life of cooling water in a water cooling system

Abstract

The present invention provides a method, device and readable medium for calculating the life of cooling water in a water cooling system. The method comprises: collecting temperature data and conductivity data of cooling water in the water cooling system within a preset time period; determining a target calculation strategy according to the temperature data and conductivity data of cooling water; wherein the target calculation strategy is a method for calculating the life of cooling water; performing characteristic value calculation on the temperature data and conductivity data of cooling water to obtain parameter characteristic values ​​of cooling water; and calculating the life of cooling water in the water cooling system according to the parameter characteristic values ​​of cooling water and the target calculation strategy. This solution can achieve accurate maintenance of the water cooling system by calculating the life of cooling water.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a method, device and readable medium for calculating the life of cooling water in a water cooling system. Background Art

[0002] The water cooling system is a device that uses cooling water as a heat exchange medium to dissipate heat and cool the body. It has been widely used in electrical, electronic, and mechanical fields due to its advantages such as high heat dissipation efficiency, low vibration, and low noise.

[0003] However, when the conductivity of the cooling water in the water cooling system increases, it will cause the risk of short circuit or even damage to the system. Factors such as oxidation caused by long-term contact between the cooling water and the circulation pipe, poor sealing of the cooling water container, and failure of the deionization exchange device will all lead to increased conductivity of the cooling water. Therefore, determining the life of the cooling water in the water cooling system and further timely maintenance of the cooling system are important for the normal operation of the water cooling system. Summary of the invention

[0004] The present invention provides a method, a device and a readable medium for calculating the life of cooling water in a water cooling system, which can achieve accurate maintenance of the water cooling system by calculating the life of the cooling water.

[0005] In a first aspect, an embodiment of the present invention provides a method for calculating the life of cooling water in a water cooling system, the method comprising:

[0006] Collect temperature data and conductivity data of cooling water in the water cooling system within a preset time period;

[0007] Determining a target calculation strategy according to the temperature data and the conductivity data of the cooling water; wherein the target calculation strategy is a method for calculating the life of the cooling water;

[0008] Performing characteristic value calculation on the temperature data and conductivity data of the cooling water to obtain parameter characteristic values ​​of the cooling water;

[0009] The life of the cooling water in the water cooling system is calculated according to the parameter characteristic value of the cooling water and the target calculation strategy.

[0010] In a possible implementation, determining a target calculation strategy according to the temperature data and the conductivity data of the cooling water includes:

[0011] Determining whether a preset sampling and analysis condition is met according to the temperature data and conductivity data of the cooling water;

[0012] If the temperature data and conductivity data of the cooling water do not meet the preset sampling and analysis conditions, the fuzzy reasoning strategy is determined as the target calculation strategy;

[0013] If the temperature data and conductivity data of the cooling water meet the preset sampling and analysis conditions, the conductivity model calculation strategy is determined as the target calculation strategy.

[0014] In a possible implementation, when the target calculation strategy is a fuzzy inference strategy, the characteristic value calculation is performed on the temperature data and the conductivity data of the cooling water to obtain the parameter characteristic value of the cooling water, including:

[0015] The mean values ​​and extreme values ​​of the temperature data and the conductivity data are calculated respectively to obtain the temperature mean value, the temperature extreme value, the conductivity mean value and the conductivity extreme value.

[0016] In a possible implementation, calculating the life of the cooling water in the water cooling system according to the parameter characteristic value of the cooling water and the target calculation strategy includes:

[0017] The temperature mean, the temperature extreme, the conductivity mean and the conductivity extreme are used as inputs for fuzzy processing to obtain an inference result;

[0018] According to the inference result, it is determined whether the life of the cooling water reaches a maintenance threshold.

[0019] In a possible implementation, when the target calculation strategy is a conductivity model calculation strategy, performing characteristic value calculation on the temperature data and conductivity data of the cooling water to obtain the parameter characteristic value of the cooling water includes:

[0020] The mean and variance of the temperature data and the conductivity data are calculated respectively to obtain the temperature mean, temperature variance, conductivity mean and conductivity variance.

[0021] In a possible implementation, calculating the life of the cooling water in the water cooling system according to the parameter characteristic value of the cooling water and the target calculation strategy includes:

[0022] The electrolyte distribution calculation model of the cooling water is obtained by training in advance through a neural network; wherein the electrolyte distribution calculation model is obtained by taking the conductivity characteristics and temperature characteristics of the cooling water collected historically as input and the expectation and standard deviation of the electrolyte concentration in the cooling water as output; the conductivity characteristics include the mean and variance of the conductivity, and the temperature characteristics include the mean and variance of the temperature;

[0023] Inputting the temperature mean, the temperature variance, the conductivity mean and the conductivity variance into the electrolyte distribution calculation model to obtain the expected and standard deviation of the electrolyte concentration in the cooling water;

[0024] Acquire at least one electrolyte concentration failure curve from a historical database; wherein the electrolyte concentration failure curve is used to characterize the change state of electrolyte concentration in cooling water over time;

[0025] For each of the electrolyte concentration failure curves, the expected value and standard deviation of the electrolyte concentration and the electrolyte concentration failure curve are used to perform maximum probability estimation to obtain at least one cooling water life calculation result.

[0026] In a possible implementation, the using the expectation and standard deviation of the electrolyte concentration and the electrolyte concentration failure curve to perform maximum probability estimation includes:

[0027] Determining a probability distribution of electrolyte concentration using the mean and standard deviation of the electrolyte concentration;

[0028] Using a sliding window of a preset size, single-step sliding sampling is performed from the initial position to the end position of the electrolyte concentration failure curve to obtain at least one set of sample sets; wherein each set of sample sets is obtained by sampling from only one position;

[0029] For each sample set, calculating the probability of the sample set appearing in the probability distribution of the electrolyte concentration;

[0030] The sliding window position corresponding to the sample set with the highest probability in the calculated results is determined as the target position;

[0031] The distance between the target position and the failure point is calculated to obtain the life of the cooling water in the water cooling system.

[0032] In a second aspect, an embodiment of the present invention provides a device for calculating the life of cooling water in a water cooling system, the device comprising:

[0033] A data acquisition module for collecting temperature data and conductivity data of cooling water in the water cooling system within a preset time period;

[0034] a strategy determination module, used to determine a target calculation strategy according to the temperature data and conductivity data of the cooling water collected by the data collection module; wherein the target calculation strategy is a method for calculating the life of the cooling water;

[0035] an eigenvalue calculation module, used for performing eigenvalue calculation on the temperature data and conductivity data of the cooling water collected by the data collection module to obtain parameter eigenvalues ​​of the cooling water;

[0036] A cooling water life calculation module is used to calculate the life of the cooling water in the water cooling system according to the parameter characteristic value of the cooling water obtained by the characteristic value calculation module and the target calculation strategy determined by the strategy determination module.

[0037] In a possible implementation, when the strategy determination module determines the target calculation strategy according to the temperature data and the conductivity data of the cooling water, it is configured to perform the following operations:

[0038] Determining whether a preset sampling and analysis condition is met according to the temperature data and conductivity data of the cooling water;

[0039] If the temperature data and conductivity data of the cooling water do not meet the preset sampling and analysis conditions, the fuzzy reasoning strategy is determined as the target calculation strategy;

[0040] If the temperature data and conductivity data of the cooling water meet the preset sampling and analysis conditions, the conductivity model calculation strategy is determined as the target calculation strategy.

[0041] In a possible implementation, when the target calculation strategy is a fuzzy inference strategy, the characteristic value calculation module performs characteristic value calculation on the temperature data and conductivity data of the cooling water, and is configured to perform the following operations when obtaining the parameter characteristic value of the cooling water:

[0042] The mean values ​​and extreme values ​​of the temperature data and the conductivity data are calculated respectively to obtain the temperature mean value, the temperature extreme value, the conductivity mean value and the conductivity extreme value.

[0043] In a possible implementation, the cooling water life calculation module is configured to perform the following operations when calculating the life of the cooling water in the water cooling system according to the parameter characteristic value of the cooling water and the target calculation strategy:

[0044] The temperature mean, the temperature extreme, the conductivity mean and the conductivity extreme are used as inputs for fuzzy processing to obtain an inference result;

[0045] According to the inference result, it is determined whether the life of the cooling water reaches a maintenance threshold.

[0046] In a possible implementation, when the target calculation strategy is the conductivity model calculation strategy, the characteristic value calculation module performs characteristic value calculation on the temperature data and the conductivity data of the cooling water to obtain the parameter characteristic value of the cooling water, and is configured to perform the following operations:

[0047] The mean and variance of the temperature data and the conductivity data are calculated respectively to obtain the temperature mean, temperature variance, conductivity mean and conductivity variance.

[0048] In a possible implementation, the cooling water life calculation module includes:

[0049] A model training unit, used for pre-training the electrolyte distribution calculation model of the cooling water through a neural network; wherein the electrolyte distribution calculation model is obtained by taking the conductivity characteristics and temperature characteristics of the cooling water collected historically as input and the expectation and standard deviation of the electrolyte concentration in the cooling water as output; the conductivity characteristics include the mean and variance of the conductivity, and the temperature characteristics include the mean and variance of the temperature;

[0050] an electrolyte concentration calculation unit, used for inputting the temperature mean, the temperature variance, the conductivity mean and the conductivity variance into the electrolyte distribution calculation model obtained by the model training unit to obtain the expected and standard deviation of the electrolyte concentration in the cooling water;

[0051] A failure curve acquisition unit, used to acquire at least one electrolyte concentration failure curve from a historical database; wherein the electrolyte concentration failure curve is used to characterize the change state of the electrolyte concentration in the cooling water over time;

[0052] A cooling water life calculation unit is used to perform maximum probability estimation for each electrolyte concentration failure curve obtained by the failure curve acquisition unit, using the expected value and standard deviation of the electrolyte concentration obtained by the electrolyte concentration calculation unit and the electrolyte concentration failure curve, to obtain at least one cooling water life calculation result.

[0053] In a possible implementation, the cooling water life calculation unit is configured to perform the following operations when performing maximum probability estimation using the expected value and standard deviation of the electrolyte concentration and the electrolyte concentration failure curve:

[0054] Determining a probability distribution of electrolyte concentration using the mean and standard deviation of the electrolyte concentration;

[0055] Using a sliding window of a preset size, single-step sliding sampling is performed from the initial position to the end position of the electrolyte concentration failure curve to obtain at least one set of sample sets; wherein each set of sample sets is obtained by sampling from only one position;

[0056] For each sample set, calculating the probability of the sample set appearing in the probability distribution of the electrolyte concentration;

[0057] The sliding window position corresponding to the sample set with the highest probability in the calculated results is determined as the target position;

[0058] The distance between the target position and the failure point is calculated to obtain the life of the cooling water in the water cooling system.

[0059] In a third aspect, an embodiment of the present invention further provides a computing device, including: at least one memory and at least one processor;

[0060] The at least one memory is used to store a machine-readable program;

[0061] The at least one processor is used to call the machine-readable program to execute any method described in the first aspect.

[0062] In a fourth aspect, an embodiment of the present invention further provides a computer-readable medium, on which computer instructions are stored. When the computer instructions are executed by a processor, the processor executes any one of the methods described in the first aspect.

[0063] It can be seen from the above technical scheme that when calculating the life of cooling water in a water cooling system, the temperature data and conductivity data of the cooling water in the water cooling system within a certain time period are first collected, and then the target calculation strategy is determined based on the collected data, that is, which cooling water life calculation method is more appropriate, and then the collected data is subjected to characteristic value calculation to obtain the parameter characteristic value of the cooling water, so that the life of the cooling water can be determined using the parameter characteristic value and the determined target calculation strategy. It can be seen that this scheme takes into account that the cooling water in the water cooling system may exist in different states, so by adopting different life calculation strategies for different states, the calculated life can be made more accurate. In addition, since temperature will affect conductivity, this scheme not only uses conductivity data as the basic data for calculating the life of cooling water, but also uses temperature data as the basic data for calculating the life of cooling water, so that the influence of temperature on conductivity can be fully taken into account, so as to achieve the purpose of improving the accuracy of the cooling water life calculation results, and then realize the precise maintenance of the water cooling system. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0065] Figure 1 It is a flow chart of a method for calculating the life of cooling water in a water cooling system provided by one embodiment of the present invention;

[0066] Figure 2is a flow chart of a method for determining a target computing strategy provided by an embodiment of the present invention;

[0067] Figure 3 is a flow chart of a method for calculating the life of cooling water in a water cooling system provided by one embodiment of the present invention;

[0068] Figure 4 is a flow chart of a maximum probability estimation method provided by an embodiment of the present invention;

[0069] Figure 5 is a schematic diagram of calculating cooling water life provided by an embodiment of the present invention;

[0070] Figure 6 is a schematic diagram of a device for calculating the life of cooling water in a water cooling system provided by one embodiment of the present invention;

[0071] Figure 7 is a schematic diagram of a cooling water life calculation module provided by an embodiment of the present invention;

[0072] Figure 8 It is a schematic diagram of a computing device provided by an embodiment of the present invention.

[0073] Reference numerals list

[0074] 101: Collect the temperature and conductivity data of the cooling water in the water cooling system within a preset time period

[0075] 102: Determine the target calculation strategy based on the temperature and conductivity data of the cooling water

[0076] 103: Calculate the characteristic values ​​of the temperature data and conductivity data of the cooling water to obtain the characteristic values ​​of the parameters of the cooling water. 104: Calculate the life of the cooling water in the water cooling system according to the characteristic values ​​of the parameters of the cooling water and the target calculation strategy.

[0077] 201: Determine whether the preset sampling and analysis conditions are met based on the temperature and conductivity data of the cooling water

[0078] 202: If the temperature data and conductivity data of the cooling water do not meet the pre-set sampling and analysis conditions, the fuzzy reasoning strategy is determined as the target calculation strategy.

[0079] 203: If the temperature data and conductivity data of the cooling water meet the pre-set sampling and analysis conditions, the conductivity model calculation strategy is determined as the target calculation strategy.

[0080] 301: Pre-train the calculation model of electrolyte distribution of cooling water through neural network

[0081] 302: Input the temperature mean, temperature variance, conductivity mean and conductivity variance into the electrolyte distribution calculation model to obtain the expected and standard deviation of the electrolyte concentration in the cooling water

[0082] 303: Get at least one electrolyte concentration failure curve from the historical database

[0083] 304: For each electrolyte concentration failure curve, use the expected value and standard deviation of the electrolyte concentration and the electrolyte concentration failure curve to perform maximum probability estimation and obtain at least one cooling water life calculation result.

[0084] 401: Using the expectation and standard deviation of electrolyte concentration, determine the probability distribution of electrolyte concentration

[0085] 402: Using a sliding window of a preset size to perform single-step sliding sample collection from the initial position to the end position of the electrolyte concentration failure curve, at least one set of sample sets is obtained.

[0086] 403: For each sample set, calculate the probability of the sample set appearing in the probability distribution of the electrolyte concentration 404: Determine the sliding window position corresponding to the sample set with the largest probability in the calculated results as the target position 405: Calculate the distance between the target position and the failure point to obtain the life of the cooling water in the water cooling system

[0087] 601: Data collection module 602: Strategy determination module 603: Feature value calculation module

[0088] 604: Cooling water life calculation module 6041: Model training unit 6042: Electrolyte concentration calculation unit 6043: Failure curve acquisition unit 6044: Cooling water life calculation unit

[0089] 100: Method for calculating the life of cooling water in a water cooling system 600: Device for calculating the life of cooling water in a water cooling system 801: Memory 802: Processor 800: Computing device DETAILED DESCRIPTION

[0090] As mentioned above, the water cooling system uses circulating water cooling, and its main power components are cooled by deionized water (cooling water) as the heat exchange medium. However, factors such as oxidation caused by long-term contact between cooling water and circulating pipes, poor sealing of cooling water containers, and failure of deionized exchange devices may cause the conductivity of deionized water to increase, bringing breakdown and short circuit risks to the water cooling system. Therefore, in order to avoid this risk, it is usually necessary to estimate the life of cooling water as the basis for planned maintenance.

[0091] At present, when judging whether the water cooling system needs maintenance, the method usually adopted is to monitor the conductivity of the cooling water in real time. When the conductivity reaches a certain value, an alarm will be issued, so that the maintenance personnel can maintain the cooling system accordingly. However, this method is a passive maintenance method, and the cooling system can only be passively maintained after the alarm occurs. This may cause additional losses due to failures. For example, when the alarm is not handled in time, the system may trip and shut down. If a fixed maintenance cycle is adopted, over-maintenance may occur. In addition, in the early stage of the water cooling system, the electrolyte in the cooling water has not reached a stable state, the sampling variance of its conductivity will be large, and the temperature will also affect the conductivity, resulting in the above-mentioned passive maintenance method through alarms. It is easy to generate false alarms. Therefore, it is often difficult to achieve accurate maintenance of the water cooling system using the above method.

[0092] Based on this, this solution considers using both the temperature data and conductivity data of the cooling water as basic data to calculate the life of the cooling water, and to achieve accurate maintenance of the water cooling system through the life of the cooling water. In addition, the different states of the cooling water are determined by the collected temperature data and conductivity data, and then the accuracy of the cooling water life is improved by using a calculation method that is more suitable for the state, so as to further improve the accuracy of the maintenance of the water cooling system.

[0093] The following is a detailed description of a method for calculating the life of cooling water in a water cooling system provided by an embodiment of the present invention, a device and a readable medium in conjunction with the accompanying drawings.

[0094] like Figure 1 As shown, the present invention provides a method 100 for calculating the life of cooling water in a water cooling system. The method may include the following steps:

[0095] Step 101: collecting temperature data and conductivity data of cooling water in a water cooling system within a preset time period;

[0096] Step 102: determining a target calculation strategy according to the temperature data and the conductivity data of the cooling water; wherein the target calculation strategy is a method for calculating the life of the cooling water;

[0097] Step 103: Calculate characteristic values ​​of the temperature data and conductivity data of the cooling water to obtain parameter characteristic values ​​of the cooling water;

[0098] Step 104: Calculate the life of the cooling water in the water cooling system according to the parameter characteristic values ​​of the cooling water and the target calculation strategy.

[0099] When calculating the life of cooling water in a water cooling system, first collect the temperature data and conductivity data of the cooling water in the water cooling system within a certain time period, and then determine the target calculation strategy based on the collected data, that is, which cooling water life calculation method is more appropriate, and then perform characteristic value calculation on the collected data to obtain the parameter characteristic value of the cooling water. In this way, the life of the cooling water can be determined using the parameter characteristic value and the determined target calculation strategy. It can be seen that this scheme takes into account that the cooling water in the water cooling system will exist in different states, so by adopting different life calculation strategies for different states, the calculated life can be made more accurate. In addition, since temperature will affect conductivity, this scheme not only uses conductivity data as the basic data for calculating the life of cooling water, but also uses temperature data as the basic data for calculating the life of cooling water. In this way, the influence of temperature on conductivity can be fully taken into account, so as to achieve the purpose of improving the accuracy of the cooling water life calculation results, and then realize the precise maintenance of the water cooling system.

[0100] In the embodiment of the present invention, by installing a temperature sensor and a device for detecting conductivity in the water cooling system, the water cooling system is monitored and detected in real time, and the collected data is stored in a database. Of course, the data stored in the database can be deleted regularly, for example, only the data within the past week is retained, and the data before one week is cleared, so as to release the invalid occupation of the memory and improve the execution efficiency of the processor. When the temperature data and conductivity data of the cooling water in the water cooling system within a preset time period are collected in step 101, the stored temperature data and conductivity data are obtained from the database. The data obtained can be all the data of the water cooling system in the database. Of course, the collection cycle can also be set, for example, the temperature data and conductivity data of the past week or the past three days are collected, so as to obtain data that can better characterize the current cooling water state.

[0101] Since the state of cooling water in the water cooling system is not always in a stable state, different situations may cause the cooling water to be in different states. For example, when the water cooling system is put into use in the early stage, the electrolyte may not reach a stable state due to reasons such as the electrolyte not being precipitated, and the variance of the conductivity in the data collected at this time will be relatively large. Therefore, the conductivity data at this time is not enough to effectively characterize the life of the cooling water. For another example, the continuous recording time obtained by real-time monitoring is too short, and the data is not enough to support data analysis. Therefore, if the above-mentioned situations are regarded as normal situations and the same calculation strategy is used to calculate the life of the cooling water, a large error is bound to occur. For example, when the water cooling system is just put into use, the electrolyte of the cooling water has not reached a stable state. If the normal way of calculating the life of the cooling water is used at this time, then it is very likely that the result obtained is that the cooling water in the water cooling system does not meet the standard and the water cooling system needs to be maintained, but it is obvious that the water cooling system has just been put into use. Based on this, in the embodiment of the present invention, these situations are fully considered, and different cooling water life calculation methods are respectively adopted for different situations, so as to obtain more accurate results through more reasonable calculation methods, so as to perform more accurate maintenance on the water cooling system. Specifically, in a possible implementation, Figure 2 As shown, step 102 can be implemented by determining the target calculation strategy according to the temperature data and conductivity data of the cooling water through the following steps:

[0102] Step 201: determining whether a preset sampling and analysis condition is met according to the temperature data and conductivity data of the cooling water;

[0103] Step 202: If the temperature data and the conductivity data of the cooling water do not meet the preset sampling and analysis conditions, the fuzzy reasoning strategy is determined as the target calculation strategy;

[0104] Step 203: If the temperature data and conductivity data of the cooling water meet the preset sampling and analysis conditions, the conductivity model calculation strategy is determined as the target calculation strategy.

[0105] In an embodiment of the present invention, when determining the target calculation strategy based on the temperature data and conductivity data of the cooling water, firstly, it is judged whether the collected temperature data and conductivity data meet the pre-set sampling and analysis conditions. When the conditions are met, the conductivity calculation model can be considered as the target calculation strategy, that is, the conductivity calculation model is used to accurately calculate the life of the cooling water, so as to perform accurate water cooling system maintenance. When the conditions are not met, fuzzy reasoning can be considered as the target calculation strategy, so that the results are output through fuzzy reasoning, and then the user is prompted to perform water cooling system maintenance through early warning. In this way, this embodiment adopts different strategies for different situations to improve the accuracy of life calculation and achieve the purpose of accurate maintenance of the water cooling system.

[0106] As can be seen from the above, in the embodiment of the present invention, when calculating the life of the cooling water in the water cooling system according to the parameter characteristics of the cooling water and the target calculation strategy in step 104, the following two target calculation strategies can be mainly adopted for different cooling water states:

[0107] The first target calculation strategy: fuzzy reasoning strategy;

[0108] The second target calculation strategy: conductivity model calculation strategy.

[0109] In the first target calculation strategy, when the temperature data and conductivity data of the cooling water do not meet the pre-set sampling and analysis conditions, that is, when the target calculation strategy is a fuzzy reasoning strategy, before calculating the life of the cooling water in the water cooling system, it is necessary to first calculate the characteristic value of the temperature data and conductivity data of the cooling water. In step 103, the characteristic value of the temperature data and conductivity data of the cooling water is calculated to obtain the parameter characteristic value of the cooling water. Specifically, it can be achieved by calculating the mean and extreme value of the temperature data and the conductivity data respectively to obtain the temperature mean, temperature extreme, conductivity mean and conductivity extreme. In this way, by using the characteristic values ​​of the temperature mean, temperature extreme, conductivity mean, conductivity extreme, etc. that can characterize the mean and extreme change of the temperature and conductivity of the cooling water, the calculation is performed, so that when the life of the cooling water is inferred by the fuzzy reasoning strategy, the inference result can be obtained more accurately.

[0110] When the cooling water life is calculated by using the calculated temperature mean, temperature extreme, conductivity mean and conductivity extreme, that is, when the target calculation strategy is the fuzzy reasoning strategy, step 104 calculates the life of the cooling water in the cooling system according to the parameter characteristic value of the cooling water and the target calculation strategy. In a possible implementation, it can be implemented as follows:

[0111] The temperature mean, temperature extreme, conductivity mean and conductivity extreme are used as input for fuzzy processing to obtain the inference result;

[0112] Based on the inference results, determine whether the life of the cooling water reaches the maintenance threshold.

[0113] In an embodiment of the present invention, when the collected cooling water temperature data and conductivity data are insufficient to support analysis, it is possible to consider setting an early warning threshold, by taking the temperature mean, temperature extreme, conductivity mean and conductivity extreme obtained after characteristic value processing as input, and combining the pre-set reasoning rules for fuzzy processing, so as to obtain whether the early warning condition is reached according to the obtained reasoning result, and further the user can decide whether to maintain the water cooling system according to whether the early warning occurs, so as to solve the situation that some special situations such as large conductivity variance and insufficient continuous recording time of data cannot rely on the conductivity model to calculate the cooling water life, so as to improve the accuracy of water cooling system maintenance. Although the prior art also adopts the early warning method, the implementation method is different from the prior art in that the present scheme is to obtain the result of judging whether the early warning condition is obtained by processing the temperature data and the conductivity data at the same time, so as to fully consider the influence of temperature on conductivity, and the present invention adopts the fuzzy processing method to improve the maintenance accuracy. It should be noted that the operation of fuzzy processing of data has been well known to people in the same field, and will not be elaborated in detail here.

[0114] In the second target calculation strategy, when the temperature data and conductivity data of the cooling water meet the pre-set sampling and analysis conditions, that is, when the target calculation strategy is the conductivity model calculation strategy, before calculating the life of the cooling water in the water cooling system, it is also necessary to calculate the characteristic value of the temperature data and conductivity data of the cooling water. In this implementation, step 103 calculates the characteristic value of the temperature data and conductivity data of the cooling water, and when obtaining the parameter characteristic value of the cooling water, the mean and variance of the temperature data and the conductivity data can be calculated respectively to obtain the temperature mean, temperature variance, conductivity mean and conductivity variance. In this way, the characteristic values ​​of the cooling water characteristics such as the obtained temperature mean, temperature variance, conductivity mean and conductivity variance are calculated to obtain a more accurate cooling water life. Since in this implementation, the temperature data and conductivity data are sufficient to support the analysis, that is, the data such as the conductivity variance will not have a large fluctuation due to the instability of the electrolyte in the cooling water, it is considered that the mean and variance can better characterize the characteristics of the cooling water.

[0115] When the obtained temperature mean, temperature variance, conductivity mean and conductivity variance are used to calculate the life of the cooling water in the water cooling system, that is, when the target calculation strategy is the conductivity model calculation strategy, step 104 calculates the life of the cooling water in the water cooling system according to the parameter characteristic value of the cooling water and the target calculation strategy. In a possible implementation, Figure 3 As shown, this can be achieved through the following steps:

[0116] Step 301: Preliminarily train a neural network to obtain an electrolyte distribution calculation model for cooling water; wherein the electrolyte distribution calculation model is obtained by taking the conductivity characteristics and temperature characteristics of cooling water collected historically as input and taking the expectation and standard deviation of the electrolyte concentration in the cooling water as output; the conductivity characteristics include the mean and variance of the conductivity, and the temperature characteristics include the mean and variance of the temperature;

[0117] Step 302: inputting the temperature mean, temperature variance, conductivity mean and conductivity variance into the electrolyte distribution calculation model to obtain the expected and standard deviation of the electrolyte concentration in the cooling water;

[0118] Step 303: Acquire at least one electrolyte concentration failure curve from a historical database; wherein the electrolyte concentration failure curve is used to characterize the change state of the electrolyte concentration in the cooling water over time;

[0119] Step 304: For each electrolyte concentration failure curve, a maximum probability estimation is performed using the expected value and standard deviation of the electrolyte concentration and the electrolyte concentration failure curve to obtain at least one cooling water life calculation result.

[0120] Considering that the conductivity will change to a certain extent at different temperatures, and there is a certain relationship between the conductivity and the concentration of the electrolyte in the cooling water. Therefore, in an embodiment of the present invention, it is considered to use a neural network to learn and fit the distribution of the electrolyte at different temperatures. By inputting the parameter characteristics obtained through feature calculation into the trained electrolyte distribution calculation model, the temperature data and conductivity data describing the current cooling water state can be converted into the electrolyte concentration situation in the current cooling water. In this way, by obtaining the electrolyte concentration failure curve from the historical database, by matching the electrolyte concentration situation in the electrolyte concentration failure curve, the most matching position of the current electrolyte concentration in the electrolyte concentration failure curve is found. Since the electrolyte concentration failure curve is a function of the electrolyte concentration with respect to time, the life of the cooling water in the current water cooling system can be accurately determined.

[0121] When training to obtain the electrolyte distribution calculation model in step 301, a supervised model training method can be considered. Specifically, the conductivity characteristics and temperature characteristics are obtained by performing feature calculation on the conductivity and temperature data of the cooling water collected historically, and then the conductivity characteristics and temperature characteristics are used as inputs for neural network learning, and the expected and standard deviation of the electrolyte concentration in the cooling water are obtained as output training. For example, the mean and variance of the temperature data and conductivity data collected historically are respectively calculated, and the expectation and variance of the electrolyte concentration of the corresponding cooling water are measured or calculated accordingly. In this way, the mean and variance of the obtained temperature data and conductivity data are used as input, and the expectation and variance of the electrolyte concentration are used as output for model training to obtain the electrolyte distribution calculation model.

[0122] That is to say, when the temperature mean, temperature variance, conductivity mean and conductivity variance are obtained respectively according to the current cooling water temperature data and conductivity data, the mean and variance are input into the electrolyte distribution calculation model to obtain the current electrolyte distribution in the cooling water.

[0123] When obtaining the electrolyte concentration failure curve from the historical database in step 303, selection may be made based on the differences among the electrolyte concentration failure curves.

[0124] For example, when the failure curves are not very different, one failure curve can be arbitrarily selected as the failure curve for calculating the cooling water life.

[0125] For another example, when among several electrolyte concentration failure curves, there are some electrolyte concentration failure curves with little difference, and the other electrolyte concentration failure curves with large difference, then you can consider selecting a failure curve to replace the electrolyte concentration failure curve with little difference, and then use this failure curve and other electrolyte concentration failure curves with large difference as failure curves for calculating cooling water life. For example, there are 6 electrolyte concentration failure curves A, B, C, D, E and F in the database, among which B, C and F are approximately the same, and the remaining A, D and E have relatively large differences. Then, you can consider selecting any one of B, C and F (using C as an example), and then use C curve and A, D, E as electrolyte concentration failure curves for calculating cooling water life.

[0126] For another example, if the electrolyte concentration failure curves in the historical data database vary greatly, then all electrolyte concentration failure curves are used as failure curves for calculating the cooling water life, and then the time for maintaining the water cooling system is further determined based on the multiple cooling water life calculation results.

[0127] When calculating the cooling water life through multiple electrolyte concentration failure curves, multiple life calculation results will be obtained. At this time, they can be displayed to the user through the touch screen system, and the user can take corresponding maintenance measures according to their needs. For example, to be on the safe side, you can choose the result with the shortest life as the final result to reduce the probability of failure of the water cooling system. For another example, you can also consider taking the average of each result as the final calculation result.

[0128] It can be understood that the electrolyte concentration failure curve obtained from the database should be a complete failure curve, that is, it includes the change of electrolyte concentration in the entire process from the beginning to the failure of the electrolyte.

[0129] When the maximum probability estimation is performed using the expected value and standard deviation of the electrolyte concentration and the electrolyte concentration failure curve in step 304, in a possible implementation, as shown in FIG. Figure 4 As shown, this can be achieved through the following steps:

[0130] Step 401: Determine the probability distribution of electrolyte concentration using the expected value and standard deviation of electrolyte concentration;

[0131] Step 402: using a sliding window of a preset size to perform single-step sliding sample collection from the initial position to the end position of the electrolyte concentration failure curve to obtain at least one set of sample sets; wherein each set of sample sets is obtained by collecting samples from only one position;

[0132] Step 403: for each sample set, calculating the probability of the sample set appearing in the probability distribution of electrolyte concentration;

[0133] Step 404: Determine the sliding window position corresponding to the sample set with the highest probability in the calculated results as the target position;

[0134] Step 405: Calculate the distance between the target position and the failure point to obtain the life of the cooling water in the water cooling system.

[0135] When using electrolyte concentration and expectation and standard deviation to estimate the maximum probability of electrolyte concentration failure, first use the expectation and standard deviation of the electrolyte concentration to determine the probability distribution of the electrolyte concentration. For example, the expectation and standard deviation of the electrolyte concentration obtained by the electrolyte distribution calculation model are μ and σ respectively. Considering that the distribution of the electrolyte concentration can be a normal distribution, the probability distribution of the electrolyte concentration is:

[0136] After obtaining the probability distribution of electrolyte concentration, such as Figure 5The schematic diagram of cooling water life calculation shown in the figure uses a sliding window of preset size to collect samples by single-step sliding from the initial position to the end position of the electrolyte concentration failure curve. Among them, the size of the sliding window can be determined according to the time size, for example, the length of a week is used as the size of the sliding window. It can also be determined according to the number of sample data collected, for example, the size of the sliding window is used to collect 10 sample points each time. When the sample window slides, the left side of the sliding window coincides with the initial moment of the electrolyte concentration failure curve as the initial position, and the right side of the sliding window coincides with the failure point as the end position. When the sliding window slides, consider collecting data in a single-step sliding manner. For example, there are 10 sample point data, namely A, B, C, D, E, F, G, H, I and J, and the size of the sliding window is to collect 5 sample data each time. Then the sample points collected for the first time are A, B, C, D and E respectively. After sliding once, the sample points collected for the second time are B, C, D, E and F respectively. After sliding once again, the sample points collected for the third time are C, D, E, F and G respectively. And so on. Each sliding data collection obtains a set of sample sets. Of course, this is just one implementation method. In other implementation methods, sliding twice can be used as a sample data collection result. In other words, this does not constitute a specific limitation on the sliding data collection method of the sliding window.

[0137] After using the sliding window to collect samples, the probability distribution formula of the electrolyte concentration obtained in step 401 is used to calculate the probability of each sample set appearing in the probability distribution of the electrolyte concentration. For example, three sample sets are collected through the sliding window, each of which contains five sample points. Taking the first sample set as an example, the sample points it contains are x1, x2, x3, x4 and x5, then the probability of the window appearing in the above-mentioned probability distribution of the electrolyte concentration is P = f(x1) * f(x2) * f(x3) * f(x4) * f(x5), so a probability value can be obtained for each sample set, for example, the three probability values ​​are P1 = 0.42, P2 = 0.96, and P3 = 0.16. It can be further obtained that the largest of the three probability values ​​is the second group. At this time, the position of the sliding window where the second group of sample data is collected is determined as the target position. The remaining life of the cooling water can be accurately obtained by calculating the distance between the right side of the sliding window and the failure point.

[0138] like Figure 6 As shown, an embodiment of the present invention provides a device 600 for calculating the life of cooling water in a water cooling system, the device comprising:

[0139] A data acquisition module 601, used to collect temperature data and conductivity data of cooling water in the water cooling system within a preset time period;

[0140] A strategy determination module 602, used to determine a target calculation strategy based on the temperature data and conductivity data of the cooling water collected by the data collection module 601; wherein the target calculation strategy is a method for calculating the life of the cooling water;

[0141] A characteristic value calculation module 603, used to perform characteristic value calculation on the temperature data and conductivity data of the cooling water collected by the data collection module 601, to obtain the parameter characteristic value of the cooling water;

[0142] A cooling water life calculation module 604 is used to calculate the life of the cooling water in the water cooling system according to the parameter characteristic values ​​of the cooling water obtained by the characteristic value calculation module 603 and the target calculation strategy determined by the strategy determination module 602.

[0143] In a possible implementation, when determining the target calculation strategy according to the temperature data and the conductivity data of the cooling water, the strategy determination module 602 is configured to perform the following operations:

[0144] Determine whether the preset sampling and analysis conditions are met based on the temperature data and conductivity data of the cooling water;

[0145] If the temperature data and conductivity data of the cooling water do not meet the preset sampling and analysis conditions, the fuzzy reasoning strategy is determined as the target calculation strategy;

[0146] If the temperature data and conductivity data of the cooling water meet the preset sampling and analysis conditions, the conductivity model calculation strategy is determined as the target calculation strategy.

[0147] In a possible implementation, when the target calculation strategy is the fuzzy reasoning strategy, the characteristic value calculation module 603 performs characteristic value calculation on the temperature data and the conductivity data of the cooling water, and is configured to perform the following operations when obtaining the parameter characteristic value of the cooling water:

[0148] The mean and extreme values ​​of the temperature data and the conductivity data are calculated respectively to obtain the temperature mean, temperature extreme, conductivity mean and conductivity extreme.

[0149] In a possible implementation, the cooling water life calculation module 604 is configured to perform the following operations when calculating the life of the cooling water in the water cooling system according to the parameter characteristic value of the cooling water and the target calculation strategy:

[0150] The temperature mean, temperature extreme, conductivity mean and conductivity extreme are used as input for fuzzy processing to obtain the inference result;

[0151] Based on the inference results, determine whether the life of the cooling water reaches the maintenance threshold.

[0152] In a possible implementation, when the target calculation strategy is the conductivity model calculation strategy, the characteristic value calculation module 603 performs characteristic value calculation on the temperature data and the conductivity data of the cooling water, and is configured to perform the following operations when obtaining the parameter characteristic value of the cooling water:

[0153] The mean and variance of the temperature data and the conductivity data are calculated respectively to obtain the temperature mean, temperature variance, conductivity mean and conductivity variance.

[0154] In one possible implementation, Figure 7 As shown, the cooling water life calculation module 604 includes:

[0155] A model training unit 6041 is used to train the electrolyte distribution calculation model of cooling water in advance through a neural network; wherein the electrolyte distribution calculation model is obtained by taking the conductivity characteristics and temperature characteristics of cooling water collected historically as input and the expectation and standard deviation of the electrolyte concentration in the cooling water as output; the conductivity characteristics include the mean and variance of the conductivity, and the temperature characteristics include the mean and variance of the temperature;

[0156] an electrolyte concentration calculation unit 6042, for inputting the temperature mean, temperature variance, conductivity mean and conductivity variance into the electrolyte distribution calculation model obtained by the model training unit 6041, to obtain the expected and standard deviation of the electrolyte concentration in the cooling water;

[0157] A failure curve acquisition unit 6043, used to acquire at least one electrolyte concentration failure curve from the historical database; wherein the electrolyte concentration failure curve is used to characterize the change state of the electrolyte concentration in the cooling water over time;

[0158] A cooling water life calculation unit 6044 is used to obtain at least one cooling water life calculation result by performing maximum probability estimation for each electrolyte concentration failure curve obtained by the failure curve acquisition unit 6043, using the expected and standard deviation of the electrolyte concentration obtained by the electrolyte concentration calculation unit 6042 and the electrolyte concentration failure curve.

[0159] In a possible implementation, the cooling water life calculation unit 6044 is configured to perform the following operations when performing maximum probability estimation using the expected value and standard deviation of the electrolyte concentration and the electrolyte concentration failure curve:

[0160] Using the mean and standard deviation of the electrolyte concentrations, determine the probability distribution of the electrolyte concentrations;

[0161] Using a sliding window of a preset size, single-step sliding sampling is performed from the initial position to the end position of the electrolyte concentration failure curve to obtain at least one set of sample sets; wherein each set of sample sets is obtained by sampling from only one position;

[0162] For each set of sample sets, calculate the probability of the sample set appearing in the probability distribution of electrolyte concentration;

[0163] The sliding window position corresponding to the sample set with the highest probability in the calculated results is determined as the target position;

[0164] The distance between the target position and the failure point is calculated to obtain the life of the cooling water in the water cooling system.

[0165] like Figure 8 As shown, an embodiment of the present invention further provides a computing device 800, comprising: at least one memory 801 and at least one processor 802;

[0166] At least one memory 801, used to store a machine-readable program;

[0167] At least one processor 802 is coupled to at least one memory 801 and is used to call a machine-readable program to execute the method 100 for calculating the life of cooling water in a water cooling system provided in any one of the above embodiments.

[0168] The present invention also provides a computer-readable medium, on which computer instructions are stored, and when the computer instructions are executed by a processor, the processor executes the method 100 for calculating the life of cooling water in a water cooling system provided in any of the above embodiments. Specifically, a system or device equipped with a storage medium can be provided, on which a software program code for implementing the functions of any of the above embodiments is stored, and a computer (or CPU or MPU) of the system or device reads and executes the program code stored in the storage medium.

[0169] In this case, the program code itself read from the storage medium can realize the function of any one of the above-mentioned embodiments, and thus the program code and the storage medium storing the program code constitute a part of the present invention.

[0170] The storage medium embodiments for providing the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), a magnetic tape, a non-volatile memory card, and a ROM. Alternatively, the program code can be downloaded from a server computer by a communication network.

[0171] In addition, it should be clear that the functions of any of the above embodiments can be implemented not only by executing the program code read by the computer, but also by enabling an operating system operating on the computer to complete part or all of the actual operations based on instructions from the program code.

[0172] In addition, it can be understood that the program code read from the storage medium is written to a memory provided in an expansion board inserted into the computer or to a memory provided in an expansion module connected to the computer, and then based on the instructions of the program code, a CPU installed on the expansion board or expansion module is enabled to perform part or all of the actual operations, thereby realizing the functions of any of the above-mentioned embodiments.

[0173] It should be noted that not all steps and modules in the above-mentioned processes and device structure diagrams are necessary, and some steps or modules can be ignored according to actual needs. The execution order of each step is not fixed and can be adjusted as needed. The system structure described in the above-mentioned embodiments can be a physical structure or a logical structure, that is, some modules may be implemented by the same physical entity, or some modules may be implemented by multiple physical entities, or may be implemented by some components in multiple independent devices. Among them, the life calculation device of cooling water in the water cooling system and the life calculation method of cooling water in the water cooling system are based on the same inventive concept.

[0174] In the above embodiments, the hardware module can be implemented mechanically or electrically. For example, a hardware module can include permanent dedicated circuits or logic (such as special processors, FPGA or ASIC) to complete the corresponding operation. The hardware module can also include programmable logic or circuits (such as general-purpose processors or other programmable processors), which can be temporarily set by software to complete the corresponding operation. The specific implementation method (mechanical method, or dedicated permanent circuit, or temporarily set circuit) can be determined based on cost and time considerations.

[0175] The present invention is shown and described in detail above through the accompanying drawings and preferred embodiments. However, the present invention is not limited to these disclosed embodiments. Based on the above multiple embodiments, those skilled in the art can know that the code review methods in the above different embodiments can be combined to obtain more embodiments of the present invention, and these embodiments are also within the protection scope of the present invention.

Claims

1. A method for calculating the life of cooling water in a water cooling system, characterized in that: include: Collecting temperature data and conductivity data of cooling water in the water cooling system within a preset time period; Determine a target calculation strategy according to the temperature data and conductivity data of the cooling water; wherein the target calculation strategy is a method for calculating the life of the cooling water; Performing characteristic value calculation on the temperature data and conductivity data of the cooling water to obtain parameter characteristic values ​​of the cooling water; Calculating the life of cooling water in the water cooling system according to the parameter characteristic value of the cooling water and the target calculation strategy; Wherein, determining the target calculation strategy according to the temperature data and conductivity data of the cooling water includes: Determining whether a preset sampling and analysis condition is met according to the temperature data and conductivity data of the cooling water; If the temperature data and conductivity data of the cooling water do not meet the preset sampling and analysis conditions, the fuzzy reasoning strategy is determined as the target calculation strategy, and the fuzzy reasoning strategy refers to respectively calculating the mean and extreme value of the temperature data and the conductivity data to obtain the temperature mean, temperature extreme, conductivity mean and conductivity extreme, and using the temperature mean, temperature extreme, conductivity mean and conductivity extreme as inputs for fuzzy processing to obtain a reasoning result; If the temperature data and conductivity data of the cooling water meet the pre-set sampling and analysis conditions, the conductivity model calculation strategy is determined as the target calculation strategy, and the conductivity model calculation strategy refers to respectively calculating the mean and variance of the temperature data and the conductivity data to obtain the temperature mean, temperature variance, conductivity mean and conductivity variance, inputting the temperature mean, temperature variance, conductivity mean and conductivity variance into the electrolyte distribution calculation model to obtain the expected and standard deviation of the electrolyte concentration in the cooling water, and then calculating the cooling water life calculation result based on the electrolyte concentration failure curve.

2. The method according to claim 1, characterized in that: When the target calculation strategy is a fuzzy reasoning strategy, it is determined whether the life of the cooling water reaches a maintenance threshold according to the reasoning result.

3. The method according to claim 1, characterized in that When the target calculation strategy is the conductivity model calculation strategy, the calculating the life of the cooling water in the water cooling system according to the parameter characteristic value of the cooling water and the target calculation strategy further includes: The electrolyte distribution calculation model of the cooling water is obtained by training in advance through a neural network; wherein the electrolyte distribution calculation model is obtained by taking the conductivity characteristics and temperature characteristics of the cooling water collected historically as input and the expectation and standard deviation of the electrolyte concentration in the cooling water as output; the conductivity characteristics include the mean and variance of the conductivity, and the temperature characteristics include the mean and variance of the temperature; Acquire at least one electrolyte concentration failure curve from a historical database; wherein the electrolyte concentration failure curve is used to characterize the change state of electrolyte concentration in cooling water over time; For each of the electrolyte concentration failure curves, the expected value and standard deviation of the electrolyte concentration and the electrolyte concentration failure curve are used to perform maximum probability estimation to obtain at least one cooling water life calculation result.

4. The method according to claim 3, characterized in that The method of using the expectation and standard deviation of the electrolyte concentration and the electrolyte concentration failure curve to perform maximum probability estimation includes: Determining a probability distribution of electrolyte concentration using the mean and standard deviation of the electrolyte concentration; Using a sliding window of a preset size, single-step sliding sampling is performed from the initial position to the end position of the electrolyte concentration failure curve to obtain at least one set of sample sets; wherein each set of sample sets is obtained by sampling from only one position; For each sample set, calculating the probability of the sample set appearing in the probability distribution of the electrolyte concentration; The sliding window position corresponding to the sample set with the highest probability in the calculated results is determined as the target position; The distance between the target position and the failure point is calculated to obtain the life of the cooling water in the water cooling system.

5. A device for calculating the life of cooling water in a water cooling system, characterized in that: include: A data acquisition module for collecting temperature data and conductivity data of cooling water in the water cooling system within a preset time period; a strategy determination module, used to determine a target calculation strategy according to the temperature data and conductivity data of the cooling water collected by the data collection module; wherein the target calculation strategy is a method for calculating the life of the cooling water; a characteristic value calculation module, used for performing characteristic value calculation on the temperature data and conductivity data of the cooling water collected by the data collection module to obtain the parameter characteristic value of the cooling water; a cooling water life calculation module, used to calculate the life of the cooling water in the water cooling system according to the parameter characteristic value of the cooling water obtained by the characteristic value calculation module and the target calculation strategy determined by the strategy determination module; Wherein, when determining the target calculation strategy according to the temperature data and conductivity data of the cooling water, the strategy determination module is configured to perform the following operations: Determining whether a preset sampling and analysis condition is met according to the temperature data and conductivity data of the cooling water; If the temperature data and conductivity data of the cooling water do not meet the preset sampling and analysis conditions, the fuzzy reasoning strategy is determined as the target calculation strategy, and the fuzzy reasoning strategy refers to respectively calculating the mean and extreme value of the temperature data and the conductivity data to obtain the temperature mean, temperature extreme, conductivity mean and conductivity extreme, and using the temperature mean, temperature extreme, conductivity mean and conductivity extreme as inputs for fuzzy processing to obtain a reasoning result; If the temperature data and conductivity data of the cooling water meet the pre-set sampling and analysis conditions, the conductivity model calculation strategy is determined as the target calculation strategy, and the conductivity model calculation strategy refers to respectively calculating the mean and variance of the temperature data and the conductivity data to obtain the temperature mean, temperature variance, conductivity mean and conductivity variance, inputting the temperature mean, temperature variance, conductivity mean and conductivity variance into the electrolyte distribution calculation model to obtain the expected and standard deviation of the electrolyte concentration in the cooling water, and then calculating the cooling water life calculation result based on the electrolyte concentration failure curve.

6. The device according to claim 5, characterized in that When the target calculation strategy is a fuzzy reasoning strategy, it is determined whether the life of the cooling water reaches a maintenance threshold according to the reasoning result.

7. The device according to claim 5, characterized in that When the target calculation strategy is the conductivity model calculation strategy, the cooling water life calculation module includes: A model training unit, used for pre-training the electrolyte distribution calculation model of the cooling water through a neural network; wherein the electrolyte distribution calculation model is obtained by taking the conductivity characteristics and temperature characteristics of the cooling water collected historically as input and the expectation and standard deviation of the electrolyte concentration in the cooling water as output; the conductivity characteristics include the mean and variance of the conductivity, and the temperature characteristics include the mean and variance of the temperature; A failure curve acquisition unit, used to acquire at least one electrolyte concentration failure curve from a historical database; wherein the electrolyte concentration failure curve is used to characterize the change state of the electrolyte concentration in the cooling water over time; A cooling water life calculation unit is used to perform maximum probability estimation for each electrolyte concentration failure curve obtained by the failure curve acquisition unit, using the expected value and standard deviation of the electrolyte concentration obtained by the electrolyte concentration calculation unit and the electrolyte concentration failure curve, to obtain at least one cooling water life calculation result.

8. The device according to claim 7, characterized in that The cooling water life calculation unit is configured to perform the following operations when performing maximum probability estimation using the expected value and standard deviation of the electrolyte concentration and the electrolyte concentration failure curve: Determining a probability distribution of electrolyte concentration using the mean and standard deviation of the electrolyte concentration; Using a sliding window of a preset size, single-step sliding sampling is performed from the initial position to the end position of the electrolyte concentration failure curve to obtain at least one set of sample sets; wherein each set of sample sets is obtained by sampling from only one position; For each sample set, calculating the probability of the sample set appearing in the probability distribution of the electrolyte concentration; The sliding window position corresponding to the sample set with the highest probability in the calculated results is determined as the target position; The distance between the target position and the failure point is calculated to obtain the life of the cooling water in the water cooling system.

9. A computing device, characterized in that include: at least one memory and at least one processor; The at least one memory is used to store a machine-readable program; The at least one processor is configured to call the machine-readable program to execute the method according to any one of claims 1 to 4.

10. A computer readable medium, characterized in that The computer readable medium stores computer instructions, which, when executed by a processor, cause the processor to perform any one of the methods of claims 1 to 4.

Images