Fire early warning method for electrochemical energy storage system

Abstract

The invention discloses a fire early warning method for an electrochemical energy storage system, and the method comprises the steps: setting a plurality of monitoring points according to the structure parameters of the electrochemical energy storage system, and determining the related monitoring data of each monitoring point; analyzing the related monitoring data, and generating a fire risk coefficient of the corresponding monitoring point based on an analysis result and an application coefficient of a corresponding preset fire evaluation index; whether an abnormal monitoring point exists is judged based on the fire risk coefficients of all the monitoring points, if yes, corresponding early warning information is generated, and an early warning instruction is generated, so that operation monitoring of the electrochemical energy storage system is accurately achieved, the fire risk coefficients are accurately judged, the abnormal monitoring point and an abnormal position are determined, and an early warning signal is sent in time; and the out-of-control probability of the electrochemical energy storage system is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of electrochemical energy storage systems, and in particular to a fire warning method for electrochemical energy storage systems. Background Art

[0002] Electrochemical energy storage systems use electrochemical batteries as energy storage carriers and store and release cyclic electrical energy through energy storage inverters. Due to the particularity of electrochemical energy storage systems, fire warning for electrochemical energy storage systems is particularly important. Existing fire warning systems are difficult to achieve accurate and timely monitoring of the internal status of electrochemical energy storage systems, and cannot reliably determine the fire risk of electrochemical energy storage systems during long-term operation, which in turn causes the electrochemical energy storage system to become out of control. Summary of the invention

[0003] To solve the above technical problems, the present application provides a fire warning method for an electrochemical energy storage system. By setting multiple monitoring points inside the electrochemical energy storage system, the relevant monitoring data of each monitoring point is determined and analyzed, the fire risk coefficient of each monitoring point is obtained, and whether the monitoring point is abnormal is determined based on the fire risk coefficient. If it is abnormal, warning information is generated and a warning instruction is sent, so as to accurately monitor the operation of the electrochemical energy storage system and accurately judge the fire risk coefficient, determine the abnormal monitoring points and abnormal locations, send warning signals in time, and reduce the probability of out-of-control of the electrochemical energy storage system.

[0004] In some embodiments of the present application, a fire warning method for an electrochemical energy storage system is provided, comprising:

[0005] Set multiple monitoring points according to the structural parameters of the electrochemical energy storage system and determine the relevant monitoring data of each monitoring point;

[0006] Analyze the relevant monitoring data, and generate a fire risk coefficient for the corresponding monitoring point based on the analysis results and the application coefficient of the corresponding preset fire evaluation index;

[0007] Based on the fire risk coefficient of all monitoring points, it is determined whether there are abnormal monitoring points. If so, corresponding warning information and warning instructions are generated.

[0008] In some embodiments of the present application, the relevant monitoring data are analyzed, including:

[0009] Acquire the real-time monitoring data of each monitoring point, calculate the correlation degree between each real-time monitoring data and the preset fire evaluation index, and set the real-time monitoring data whose correlation degree with each preset fire evaluation index is greater than the preset correlation degree threshold as the relevant monitoring data of the corresponding monitoring point;

[0010] Establish a time reference line according to the time length of the current monitoring cycle, and set the monitoring time nodes in the current monitoring cycle based on the preset time interval;

[0011] Preset a standard monitoring data interval for the relevant monitoring data of each preset fire evaluation index, obtain multiple relevant monitoring data of each monitoring point according to each monitoring time node, map the relevant monitoring data of each monitoring time node and the critical standard monitoring data of the standard monitoring data interval of the relevant monitoring data onto a time reference line, and generate a monitoring data analysis diagram;

[0012] Generate two boundary lines corresponding to the relevant monitoring data according to the critical standard monitoring data of the standard monitoring data interval of the relevant monitoring data, and judge whether each relevant monitoring data in the monitoring data analysis diagram exceeds the corresponding boundary line;

[0013] If so, determine the excess area of ​​each relevant monitoring data that exceeds the corresponding boundary line, and obtain the first time node and the second time node adjacent to each excess area of ​​the relevant monitoring data and the corresponding boundary line. If the first time node and the second time node are both within the same monitoring time node, the corresponding excess area is eliminated. If they are not within the same monitoring time node, calculate the total excess area of ​​the remaining excess area, and set the deviation coefficient of the corresponding relevant monitoring data and the corresponding standard monitoring data interval according to the total excess area;

[0014] If not, directly set the deviation coefficient between the corresponding relevant monitoring data and the corresponding standard monitoring data interval as the preset deviation coefficient threshold;

[0015] A fire evaluation value corresponding to the preset fire evaluation index is generated according to the deviation coefficient of the relevant monitoring data of each preset fire evaluation index.

[0016] In some embodiments of the present application, the deviation coefficient between the corresponding relevant monitoring data and the corresponding standard monitoring data interval is set according to the exceeded total area, including:

[0017] Preset a first preset exceeding total area interval, a second preset exceeding total area interval, a third preset exceeding total area interval, and a fourth preset exceeding total area interval;

[0018] When the total excess area of ​​the remaining excess area of ​​the relevant monitoring data is within a first preset excess total area interval, the deviation coefficient between the corresponding relevant monitoring data and the corresponding standard monitoring data interval is set to be the first preset deviation coefficient;

[0019] When the total excess area of ​​the remaining excess region of the relevant monitoring data is within a second preset excess total area interval, the deviation coefficient between the corresponding relevant monitoring data and the corresponding standard monitoring data interval is set to be the second preset deviation coefficient;

[0020] When the total excess area of ​​the remaining excess region of the relevant monitoring data is within a third preset excess total area interval, the deviation coefficient between the corresponding relevant monitoring data and the corresponding standard monitoring data interval is set to the third preset deviation coefficient;

[0021] When the total excess area of ​​the remaining excess region of the relevant monitoring data is within a fourth preset excess total area interval, the deviation coefficient between the corresponding relevant monitoring data and the corresponding standard monitoring data interval is set to be the fourth preset deviation coefficient.

[0022] In some embodiments of the present application, generating a fire evaluation value corresponding to a preset fire evaluation indicator according to a deviation coefficient of relevant monitoring data of each preset fire evaluation indicator includes:

[0023] The calculation formula of the fire evaluation value is:

[0024]

[0025] Among them, H is the fire evaluation value corresponding to the preset fire evaluation index, n is the total number of relevant monitoring data corresponding to the preset fire evaluation index at the current monitoring point, pi is the deviation coefficient of the i-th relevant monitoring data corresponding to the preset fire evaluation index, and ai is the weight coefficient of the i-th relevant monitoring data corresponding to the preset fire evaluation index.

[0026] In some embodiments of the present application, based on the analysis result and the application coefficient of the corresponding preset fire evaluation index, a fire risk coefficient of the corresponding monitoring point is generated, including:

[0027] Obtaining application information of each preset fire evaluation index in the historical fire log, wherein the application information includes timeliness information, application information, and accuracy information;

[0028] Determine the discovery time node and judgment time node of the corresponding preset fire evaluation indicator according to the timeliness information of each preset fire evaluation indicator in each historical fire log, and compare them with the standard discovery time node and standard judgment time node in the corresponding historical fire log to obtain the first time deviation and the second time deviation of each preset fire evaluation indicator in each historical fire log, and generate the timeliness of the corresponding preset fire evaluation indicator according to the multiple first time deviations and the second time deviations;

[0029] Determining the application form of the preset fire evaluation index in the corresponding historical fire log according to the application information of each preset fire evaluation index in the historical fire log, wherein the application form includes single application and joint application;

[0030] When used alone, determine the number of times the corresponding preset fire evaluation index is used alone in all historical fire logs, and when used alone, the first influence of the corresponding preset fire evaluation index on the fire warning results of the historical fire logs;

[0031] When it is a joint application, determining the number of joint applications of the corresponding preset fire evaluation index in all historical fire logs, and the second influence of the corresponding preset fire evaluation index on the fire warning results of the historical fire logs when the joint application is performed;

[0032] Generate the applicability of corresponding preset fire assessment indicators according to the plurality of first impact degrees and second impact degrees;

[0033] Determine the similarity between the historical fire evaluation value of the preset fire evaluation index and the fire warning result of each historical fire log according to the accurate information of each preset fire evaluation index in the historical fire log, and generate the accuracy of the corresponding preset fire evaluation index according to the number of historical fire logs with similarity greater than a preset similarity threshold and the number of all historical fire logs;

[0034] Generate an application coefficient corresponding to each preset fire evaluation indicator according to the timeliness, operability and accuracy of each preset fire evaluation indicator;

[0035] The calculation formula of the application coefficient is:

[0036]

[0037] Wherein, Y is the application coefficient, k1 is the first application conversion coefficient, c1 is the weight coefficient of timeliness, t1s is the first time deviation of the preset fire evaluation index in the sth historical fire log, b1 is the weight coefficient of the first time deviation, t2s is the second time deviation of the preset fire evaluation index in the sth historical fire log, b2 is the weight coefficient of the second time deviation, m1 is the total number of historical fire logs, k2 is the second application conversion coefficient, w1 is the weight coefficient of the first influence, m2 is the number of historical fire logs in which the preset fire evaluation index is used alone in the historical fire log, q1z is the first influence of the preset fire evaluation index in the zth historical fire log used alone, w2 is the weight coefficient of the second influence, q2v is the second influence of the preset fire evaluation index in the vth historical fire log used jointly, m3 is the number of historical fire logs in which the preset fire evaluation index is used jointly in the historical fire log, k3 is the third application conversion coefficient, and m4 is the number of historical fire logs whose similarity is greater than the preset similarity threshold;

[0038] Generate a fire risk coefficient for the corresponding monitoring point based on the fire evaluation values ​​of all preset fire evaluation indicators of each monitoring point in the analysis results and the application coefficient of the corresponding preset fire evaluation indicators;

[0039] The calculation formula of the fire risk coefficient of the monitoring point is:

[0040]

[0041] Among them, F is the fire risk coefficient, Hz is the fire evaluation value of the z-th preset fire evaluation index, and Yz is the application coefficient of the z-th preset fire evaluation index.

[0042] In some embodiments of the present application, judging whether there is an abnormal monitoring point based on the fire risk coefficient of all monitoring points includes:

[0043] Presetting a first preset fire risk coefficient and a second preset fire risk coefficient;

[0044] When the fire risk coefficient of the monitoring point is less than the first preset fire risk coefficient, the corresponding monitoring point is judged to be a healthy monitoring point;

[0045] When the fire risk coefficient of the monitoring point is between the first preset fire risk coefficient and the second preset fire risk coefficient, the corresponding monitoring point is judged to be a general monitoring point;

[0046] When the fire risk coefficient of the monitoring point is greater than the second preset fire risk coefficient, the corresponding monitoring point is judged to be an abnormal monitoring point.

[0047] In some embodiments of the present application, the method further comprises:

[0048] The general monitoring points are set as characteristic monitoring points, and the data change characteristics of the relevant monitoring data corresponding to the current characteristic monitoring points in the future period are predicted based on the historical fire data of the characteristic monitoring points;

[0049] The judgment results of the current characteristic monitoring points are updated according to the data change characteristics of the relevant monitoring data in the future period.

[0050] In some embodiments of the present application, the correction coefficient of the fire risk coefficient of the current characteristic monitoring point is generated according to the predicted monitoring data of the characteristic monitoring point in the future period, including:

[0051] Obtain historical fire data of historical characteristic periods of multiple historical fire logs at each characteristic monitoring point, take the characteristic period of each historical fire log as a time reference line, set historical monitoring time nodes according to preset time intervals, collect historical fire data based on each historical monitoring time node, and map it to the corresponding time reference line to obtain a fire data change curve for each historical fire log;

[0052] Determine the fire change characteristics of the corresponding historical fire data according to the fire data change curve of each historical fire log, wherein the fire change characteristics include the historical data deviation value between the historical fire data and the standard monitoring data interval and the historical fluctuation value of the historical fire data at each historical monitoring time node;

[0053] According to the monitoring data analysis diagram of the relevant monitoring data of each characteristic monitoring point, the relevant monitoring data with a fluctuation degree greater than the preset fluctuation degree at the corresponding characteristic monitoring point is screened out, and the real-time characteristic period of the screened relevant monitoring data and the change characteristics to be judged within the real-time characteristic period are determined, and the change characteristics to be judged include the real-time data deviation value between the relevant monitoring data and the standard monitoring data interval and the real-time fluctuation value of the relevant monitoring data at each monitoring time node in the real-time characteristic period, wherein the real-time characteristic period is smaller than the historical characteristic period;

[0054] The to-be-judged change features of the selected relevant monitoring data are compared with multiple fire change features of historical fire data of the same data type, and the similarity between the to-be-judged change features and the fire change features is generated according to the comparison results;

[0055] If the similarity between the change feature to be judged and multiple fire change features is greater than the preset similarity threshold, the data change characteristics of the corresponding relevant monitoring data in the future time period are determined according to the fire change feature with the greatest similarity, and the judgment result of the characteristic monitoring point is updated to the abnormal monitoring point.

[0056] If the similarities between the change feature to be judged and multiple fire change features are all less than the preset similarity threshold, the judgment result of the feature monitoring point is updated to a healthy monitoring point.

[0057] In some embodiments of the present application, if present, generating corresponding warning information and generating warning instructions include:

[0058] If there are abnormal monitoring points, determine the location information of the abnormal monitoring points and the abnormal related monitoring data, and combine the fire risk coefficient of the abnormal monitoring points to generate corresponding warning information and send warning instructions.

[0059] Compared with the prior art, the fire warning method for an electrochemical energy storage system in the embodiment of the present application has the following beneficial effects:

[0060] By setting multiple monitoring points inside the electrochemical energy storage system, determining and analyzing the relevant monitoring data of each monitoring point, the fire risk coefficient of each monitoring point is obtained, and whether the monitoring point is abnormal is determined based on the fire risk coefficient. If it is abnormal, early warning information is generated and an early warning instruction is sent, so as to accurately monitor the operation of the electrochemical energy storage system and accurately judge the fire risk coefficient, determine the abnormal monitoring point and abnormal position, send early warning signals in time, and reduce the probability of out-of-control of the electrochemical energy storage system. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a flow chart of a fire warning method for an electrochemical energy storage system in a preferred embodiment of the present application. DETAILED DESCRIPTION

[0062] The specific implementation methods of the present application are further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present application but are not intended to limit the scope of the present application.

[0063] In the description of the present application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0064] The terms "first", "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0065] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0066] like Figure 1 As shown, a fire warning method for an electrochemical energy storage system according to an embodiment of the present application includes:

[0067] Step S101: setting a plurality of monitoring points according to the structural parameters of the electrochemical energy storage system, and determining relevant monitoring data of each monitoring point;

[0068] Step S102: Analyze the relevant monitoring data, and generate a fire risk coefficient of the corresponding monitoring point based on the analysis result and the application coefficient of the corresponding preset fire evaluation index;

[0069] Step S103: Based on the fire risk coefficients of all monitoring points, determine whether there are abnormal monitoring points. If so, generate corresponding warning information and generate warning instructions.

[0070] In this embodiment, the real-time monitoring data includes the temperature, current, voltage, generated hydrogen concentration, carbon monoxide concentration, etc. of the battery cells inside the electrochemical energy storage system, which are collected by multiple monitoring sensors at the corresponding monitoring points. The preset fire evaluation index is set according to the judgment criteria and expert judgment basis corresponding to multiple fire types in the historical fire log. Each historical monitoring log includes historical fire evaluation values ​​of the preset fire evaluation index at multiple historical monitoring time nodes.

[0071] In some embodiments of the present application, the relevant monitoring data are analyzed, including:

[0072] Acquire the real-time monitoring data of each monitoring point, calculate the correlation degree between each real-time monitoring data and the preset fire evaluation index, and set the real-time monitoring data whose correlation degree with each preset fire evaluation index is greater than the preset correlation degree threshold as the relevant monitoring data of the corresponding monitoring point;

[0073] Establish a time reference line according to the time length of the current monitoring cycle, and set the monitoring time nodes in the current monitoring cycle based on the preset time interval;

[0074] Preset a standard monitoring data interval for the relevant monitoring data of each preset fire evaluation index, obtain multiple relevant monitoring data of each monitoring point according to each monitoring time node, map the relevant monitoring data of each monitoring time node and the critical standard monitoring data of the standard monitoring data interval of the relevant monitoring data onto a time reference line, and generate a monitoring data analysis diagram;

[0075] Generate two boundary lines corresponding to the relevant monitoring data according to the critical standard monitoring data of the standard monitoring data interval of the relevant monitoring data, and judge whether each relevant monitoring data in the monitoring data analysis diagram exceeds the corresponding boundary line;

[0076] If so, determine the excess area of ​​each relevant monitoring data that exceeds the corresponding boundary line, and obtain the first time node and the second time node adjacent to each excess area of ​​the relevant monitoring data and the corresponding boundary line. If the first time node and the second time node are both within the same monitoring time node, the corresponding excess area is eliminated. If they are not within the same monitoring time node, calculate the total excess area of ​​the remaining excess area, and set the deviation coefficient of the corresponding relevant monitoring data and the corresponding standard monitoring data interval according to the total excess area;

[0077] If not, directly set the deviation coefficient between the corresponding relevant monitoring data and the corresponding standard monitoring data interval as the preset deviation coefficient threshold;

[0078] A fire evaluation value corresponding to the preset fire evaluation index is generated according to the deviation coefficient of the relevant monitoring data of each preset fire evaluation index.

[0079] In this embodiment, the preset deviation coefficient threshold refers to 0.2, which refers to the minimum deviation coefficient. The preset deviation coefficient threshold is smaller than the first preset deviation coefficient.

[0080] In some embodiments of the present application, the deviation coefficient between the corresponding relevant monitoring data and the corresponding standard monitoring data interval is set according to the exceeded total area, including:

[0081] Preset a first preset exceeding total area interval, a second preset exceeding total area interval, a third preset exceeding total area interval, and a fourth preset exceeding total area interval;

[0082] When the total excess area of ​​the remaining excess area of ​​the relevant monitoring data is within a first preset excess total area interval, the deviation coefficient between the corresponding relevant monitoring data and the corresponding standard monitoring data interval is set to be the first preset deviation coefficient;

[0083] When the total excess area of ​​the remaining excess region of the relevant monitoring data is within a second preset excess total area interval, the deviation coefficient between the corresponding relevant monitoring data and the corresponding standard monitoring data interval is set to be the second preset deviation coefficient;

[0084] When the total excess area of ​​the remaining excess region of the relevant monitoring data is within a third preset excess total area interval, the deviation coefficient between the corresponding relevant monitoring data and the corresponding standard monitoring data interval is set to the third preset deviation coefficient;

[0085] When the total excess area of ​​the remaining excess region of the relevant monitoring data is within a fourth preset excess total area interval, the deviation coefficient between the corresponding relevant monitoring data and the corresponding standard monitoring data interval is set to be the fourth preset deviation coefficient.

[0086] In this embodiment, the first preset excess total area interval < the second preset excess total area interval < the third preset excess total area interval < the fourth preset excess total area interval, and the preset deviation coefficient threshold < the first preset deviation coefficient < the second preset deviation coefficient < the third preset deviation coefficient < the fourth preset deviation coefficient.

[0087] In this embodiment, when the total area of ​​the exceeding area is larger, it means that the deviation degree between the relevant monitoring data and the corresponding marked monitoring data interval is larger, the deviation time is longer, and the fluctuation degree of the relevant monitoring data is larger, that is, the deviation coefficient of the corresponding relevant monitoring data is larger, and the fire evaluation value corresponding to the preset fire evaluation index is generated according to the deviation coefficient of each relevant monitoring data. When the deviation coefficient of the relevant monitoring data of the preset fire evaluation index is larger, it means that the risk of fire at the corresponding monitoring point assessed by the current preset fire evaluation index is greater, that is, the larger the fire evaluation value, which lays the foundation for the subsequent generation of the fire risk coefficient, improves the accuracy of the fire risk coefficient, and ensures the timeliness and effectiveness of fire warning.

[0088] In some embodiments of the present application, generating a fire evaluation value corresponding to a preset fire evaluation index according to a deviation coefficient of relevant monitoring data of each preset fire evaluation index includes:

[0089] The calculation formula of the fire evaluation value is:

[0090]

[0091] Among them, H is the fire evaluation value corresponding to the preset fire evaluation index, n is the total number of relevant monitoring data corresponding to the preset fire evaluation index at the current monitoring point, pi is the deviation coefficient of the i-th relevant monitoring data corresponding to the preset fire evaluation index, and ai is the weight coefficient of the i-th relevant monitoring data corresponding to the preset fire evaluation index.

[0092] In this embodiment, the weight coefficient is calculated based on the sum of the correlation degree between the corresponding relevant monitoring data and the corresponding preset fire evaluation index and the correlation degree of all relevant monitoring data corresponding to the preset fire evaluation index.

[0093] In this embodiment, by calculating the fire evaluation values ​​of multiple preset fire evaluation indicators for each monitoring point and combining the subsequent application coefficient of each preset fire evaluation indicator, the fire risk coefficient of each monitoring point is accurately calculated, thereby improving the credibility and accuracy of subsequent fire warning instructions.

[0094] In some embodiments of the present application, based on the analysis result and the application coefficient of the corresponding preset fire evaluation index, a fire risk coefficient of the corresponding monitoring point is generated, including:

[0095] Obtaining application information of each preset fire evaluation index in the historical fire log, wherein the application information includes timeliness information, application information, and accuracy information;

[0096] Determine the discovery time node and judgment time node of the corresponding preset fire evaluation indicator according to the timeliness information of each preset fire evaluation indicator in each historical fire log, and compare them with the standard discovery time node and standard judgment time node in the corresponding historical fire log to obtain the first time deviation and the second time deviation of each preset fire evaluation indicator in each historical fire log, and generate the timeliness of the corresponding preset fire evaluation indicator according to the multiple first time deviations and the second time deviations;

[0097] Determining the application form of the preset fire evaluation index in the corresponding historical fire log according to the application information of each preset fire evaluation index in the historical fire log, wherein the application form includes single application and joint application;

[0098] When used alone, determine the number of times the corresponding preset fire evaluation index is used alone in all historical fire logs, and when used alone, the first influence of the corresponding preset fire evaluation index on the fire warning results of the historical fire logs;

[0099] When it is a joint application, determining the number of joint applications of the corresponding preset fire evaluation index in all historical fire logs, and the second influence of the corresponding preset fire evaluation index on the fire warning results of the historical fire logs when the joint application is performed;

[0100] Generate the applicability of corresponding preset fire assessment indicators according to the plurality of first impact degrees and second impact degrees;

[0101] Determine the similarity between the historical fire evaluation value of the preset fire evaluation index and the fire warning result of each historical fire log according to the accurate information of each preset fire evaluation index in the historical fire log, and generate the accuracy of the corresponding preset fire evaluation index according to the number of historical fire logs with similarity greater than a preset similarity threshold and the number of all historical fire logs;

[0102] Generate an application coefficient corresponding to each preset fire evaluation indicator according to the timeliness, operability and accuracy of each preset fire evaluation indicator;

[0103] The calculation formula of the application coefficient is:

[0104]

[0105] Wherein, Y is the application coefficient, k1 is the first application conversion coefficient, c1 is the weight coefficient of timeliness, t1s is the first time deviation of the preset fire evaluation index in the sth historical fire log, b1 is the weight coefficient of the first time deviation, t2s is the second time deviation of the preset fire evaluation index in the Sth historical fire log, b2 is the weight coefficient of the second time deviation, m1 is the total number of historical fire logs, k2 is the second application conversion coefficient, w1 is the weight coefficient of the first influence, m2 is the number of historical fire logs in which the preset fire evaluation index is used in the historical fire log in the form of a single application, q1z is the first influence of the preset fire evaluation index in the zth historical fire log in which it is used in the form of a single application, w2 is the weight coefficient of the second influence, q2v is the second influence of the preset fire evaluation index in the vth historical fire log in which it is used in combination, m3 is the number of historical fire logs in which the preset fire evaluation index is used in the historical fire log in the form of a joint application, k3 is the third application conversion coefficient, and m4 is the number of historical fire logs whose similarity is greater than a preset similarity threshold;

[0106] Generate a fire risk coefficient for the corresponding monitoring point based on the fire evaluation values ​​of all preset fire evaluation indicators of each monitoring point in the analysis results and the application coefficient of the corresponding preset fire evaluation indicators;

[0107] The calculation formula of the fire risk coefficient of the monitoring point is:

[0108]

[0109] Among them, F is the fire risk coefficient, Hz is the fire evaluation value of the z-th preset fire evaluation index, and Yz is the application coefficient of the z-th preset fire evaluation index.

[0110] In this embodiment, the historical fire log refers to a detailed record process when a fire occurs in the electrochemical energy storage system. The discovery time node of the preset fire evaluation indicator refers to the time node when the corresponding preset fire evaluation indicator in the historical fire log finds that there is a fire risk. The judgment time node refers to the time node when the corresponding preset fire evaluation indicator in the historical fire log determines that there is a fire risk. The standard discovery time node and the standard judgment time node refer to the accurate time nodes when the fire risk is discovered and determined to exist in the corresponding historical fire log, respectively.

[0111] In this embodiment, separate use refers to a separate analysis based on the corresponding preset fire evaluation index. The first impact degree refers to the degree of influence on the fire warning results of the historical fire log when the corresponding preset fire evaluation index is analyzed separately. When used in combination, the corresponding preset fire evaluation index is analyzed together with other preset fire evaluation indicators. The second impact degree refers to the degree of influence on the fire warning results of the historical fire log when the corresponding preset fire evaluation index is analyzed together with other preset fire evaluation indicators.

[0112] In this embodiment, similarity refers to the consistency between the historical fire evaluation values ​​of the preset evaluation indicators in the historical fire logs and the corresponding fire warning results.

[0113] In this embodiment, by analyzing the timeliness, applicability and accuracy of each preset fire evaluation indicator in multiple historical fire logs, the application coefficient of each preset fire evaluation indicator, that is, the accuracy and credibility of each preset fire evaluation indicator, is determined, and the fire risk coefficient of each monitoring point is increased according to the fire evaluation value determined by the preset fire evaluation indicator and the application coefficient. It is judged whether there is an abnormality based on the fire risk coefficient, thereby improving the accuracy and timeliness of fire warning of the electrochemical energy storage system.

[0114] In some embodiments of the present application, judging whether there is an abnormal monitoring point based on the fire risk coefficient of all monitoring points includes:

[0115] Presetting a first preset fire risk coefficient and a second preset fire risk coefficient;

[0116] When the fire risk coefficient of the monitoring point is less than the first preset fire risk coefficient, the corresponding monitoring point is judged to be a healthy monitoring point;

[0117] When the fire risk coefficient of the monitoring point is between the first preset fire risk coefficient and the second preset fire risk coefficient, the corresponding monitoring point is judged to be a general monitoring point;

[0118] When the fire risk coefficient of the monitoring point is greater than the second preset fire risk coefficient, the corresponding monitoring point is judged to be an abnormal monitoring point.

[0119] In this embodiment, the first preset fire risk coefficient is less than the second preset fire risk coefficient, a healthy monitoring point refers to a monitoring point that has no fire risk in the future period, a general monitoring point refers to a monitoring point that has no fire risk at the current monitoring point, and an abnormal monitoring point refers to a monitoring point that has a fire risk at the current monitoring point.

[0120] In some embodiments of the present application, the method further comprises:

[0121] The general monitoring points are set as characteristic monitoring points, and the data change characteristics of the relevant monitoring data corresponding to the current characteristic monitoring points in the future period are predicted based on the historical fire data of the characteristic monitoring points;

[0122] The judgment results of the current characteristic monitoring points are updated according to the data change characteristics of the relevant monitoring data in the future period.

[0123] In this embodiment, when the similarity between the real-time change characteristics of the relevant monitoring data and the fire change characteristics of the historical fire data is greater than the preset similarity threshold, the judgment result of the corresponding characteristic monitoring point is updated to an abnormal monitoring point and a warning signal is sent. If the similarities are less than the preset similarity threshold, the judgment result of the corresponding characteristic monitoring point is updated to a healthy monitoring point.

[0124] In some embodiments of the present application, the correction coefficient of the fire risk coefficient of the current characteristic monitoring point is generated according to the predicted monitoring data of the characteristic monitoring point in the future period, including:

[0125] Obtain historical fire data of historical characteristic periods of multiple historical fire logs at each characteristic monitoring point, take the characteristic period of each historical fire log as a time reference line, set historical monitoring time nodes according to preset time intervals, collect historical fire data based on each historical monitoring time node, and map it to the corresponding time reference line to obtain a fire data change curve for each historical fire log;

[0126] Determine the fire change characteristics of the corresponding historical fire data according to the fire data change curve of each historical fire log, wherein the fire change characteristics include the historical data deviation value between the historical fire data and the standard monitoring data interval and the historical fluctuation value of the historical fire data at each historical monitoring time node;

[0127] According to the monitoring data analysis diagram of the relevant monitoring data of each characteristic monitoring point, the relevant monitoring data with a fluctuation degree greater than the preset fluctuation degree at the corresponding characteristic monitoring point is screened out, and the real-time characteristic period of the screened relevant monitoring data and the change characteristics to be judged within the real-time characteristic period are determined, and the change characteristics to be judged include the real-time data deviation value between the relevant monitoring data and the standard monitoring data interval and the real-time fluctuation value of the relevant monitoring data at each monitoring time node in the real-time characteristic period, wherein the real-time characteristic period is smaller than the historical characteristic period;

[0128] The to-be-judged change features of the selected relevant monitoring data are compared with multiple fire change features of historical fire data of the same data type, and the similarity between the to-be-judged change features and the fire change features is generated according to the comparison results;

[0129] If the similarity between the change feature to be judged and multiple fire change features is greater than the preset similarity threshold, the data change characteristics of the corresponding relevant monitoring data in the future time period are determined according to the fire change feature with the greatest similarity, and the judgment result of the characteristic monitoring point is updated to the abnormal monitoring point.

[0130] If the similarities between the change feature to be judged and multiple fire change features are all less than the preset similarity threshold, the judgment result of the feature monitoring point is updated to a healthy monitoring point.

[0131] In this embodiment, the historical characteristic period refers to the period from the time node when the historical fire data in the historical fire log fluctuated greatly to the time when the fire occurred. The historical fire data refers to the monitoring data that changed greatly when a fire occurred. The same data type as the historical fire data refers to the monitoring data obtained by the same monitoring sensor. The real-time characteristic period refers to the time node from which the relevant monitoring data fluctuated greatly in the current monitoring cycle to the latest monitoring time node.

[0132] In this embodiment, the similarity refers to the similarity of the difference between the real-time data deviation value and the historical data deviation value of the relevant monitoring data of the monitoring time node of each current monitoring cycle and the historical monitoring time node and the corresponding historical fire data, and the similarity of the fluctuation value of the real-time fluctuation value and the historical fluctuation value of the relevant monitoring data of the monitoring time node of each current monitoring cycle and the historical monitoring time node and the corresponding historical fire data.

[0133] In this embodiment, the data change characteristics of the future time period are determined based on the real-time change characteristics and the fire change characteristics with the maximum similarity. The similar parts of the real-time change characteristics are screened out from the fire change characteristics with the maximum similarity. The fire change characteristics after the similar parts are the data change characteristics of the relevant monitoring data in the future time period.

[0134] In this embodiment, by predicting the data change characteristics of the future time period of the relevant monitoring data with large fluctuations, the accuracy of the judgment result of each monitoring point is improved, abnormalities are discovered and warned in time, and the accuracy of fire warning of the electrochemical energy storage system is improved.

[0135] In some embodiments of the present application, if present, generating corresponding warning information and generating warning instructions include:

[0136] If there are abnormal monitoring points, determine the location information of the abnormal monitoring points and the abnormal related monitoring data, and combine the fire risk coefficient of the abnormal monitoring points to generate corresponding warning information and send warning instructions.

[0137] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and substitutions can be made without departing from the technical principles of the present application. These improvements and substitutions should also be regarded as the scope of protection of the present application.

Claims

1. A fire warning method for an electrochemical energy storage system, characterized in that: include: Setting multiple monitoring points according to the structural parameters of the electrochemical energy storage system and determining the relevant monitoring data of each monitoring point; Analyze the relevant monitoring data, and generate a fire risk coefficient for the corresponding monitoring point based on the analysis results and the application coefficient of the corresponding preset fire evaluation index; Based on the fire risk coefficient of all monitoring points, it is determined whether there are abnormal monitoring points. If so, corresponding warning information and warning instructions are generated.

2. The electrochemical energy storage system fire warning method according to claim 1, characterized in that: Analyze relevant monitoring data, including: Acquire the real-time monitoring data of each monitoring point, calculate the correlation degree between each real-time monitoring data and the preset fire evaluation index, and set the real-time monitoring data whose correlation degree with each preset fire evaluation index is greater than the preset correlation degree threshold as the relevant monitoring data of the corresponding monitoring point; Establish a time reference line according to the time length of the current monitoring cycle, and set the monitoring time nodes in the current monitoring cycle based on the preset time interval; Preset a standard monitoring data interval for the relevant monitoring data of each preset fire evaluation index, obtain multiple relevant monitoring data of each monitoring point according to each monitoring time node, map the relevant monitoring data of each monitoring time node and the critical standard monitoring data of the standard monitoring data interval of the relevant monitoring data onto a time reference line, and generate a monitoring data analysis diagram; Generate two boundary lines corresponding to the relevant monitoring data according to the critical standard monitoring data of the standard monitoring data interval of the relevant monitoring data, and judge whether each relevant monitoring data in the monitoring data analysis diagram exceeds the corresponding boundary line; If so, determine the excess area of ​​each relevant monitoring data that exceeds the corresponding boundary line, and obtain the first time node and the second time node adjacent to each excess area of ​​the relevant monitoring data and the corresponding boundary line. If the first time node and the second time node are both within the same monitoring time node, the corresponding excess area is eliminated. If they are not within the same monitoring time node, calculate the total excess area of ​​the remaining excess area, and set the deviation coefficient of the corresponding relevant monitoring data and the corresponding standard monitoring data interval according to the total excess area; If not, directly set the deviation coefficient between the corresponding relevant monitoring data and the corresponding standard monitoring data interval as the preset deviation coefficient threshold; A fire evaluation value corresponding to the preset fire evaluation index is generated according to the deviation coefficient of the relevant monitoring data of each preset fire evaluation index.

3. The fire warning method for an electrochemical energy storage system according to claim 2, characterized in that: The deviation coefficient between the corresponding relevant monitoring data and the corresponding standard monitoring data interval is set according to the total area exceeded, including: Preset a first preset exceeding total area interval, a second preset exceeding total area interval, a third preset exceeding total area interval, and a fourth preset exceeding total area interval; When the total excess area of ​​the remaining excess area of ​​the relevant monitoring data is within a first preset excess total area interval, the deviation coefficient between the corresponding relevant monitoring data and the corresponding standard monitoring data interval is set to be the first preset deviation coefficient; When the total excess area of ​​the remaining excess region of the relevant monitoring data is within a second preset excess total area interval, the deviation coefficient between the corresponding relevant monitoring data and the corresponding standard monitoring data interval is set to be the second preset deviation coefficient; When the total excess area of ​​the remaining excess region of the relevant monitoring data is within a third preset excess total area interval, the deviation coefficient between the corresponding relevant monitoring data and the corresponding standard monitoring data interval is set to the third preset deviation coefficient; When the total excess area of ​​the remaining excess region of the relevant monitoring data is within a fourth preset excess total area interval, the deviation coefficient between the corresponding relevant monitoring data and the corresponding standard monitoring data interval is set to be the fourth preset deviation coefficient.

4. The fire warning method for an electrochemical energy storage system according to claim 3, characterized in that: The fire evaluation value corresponding to each preset fire evaluation index is generated according to the deviation coefficient of the relevant monitoring data of each preset fire evaluation index, including: The calculation formula of the fire evaluation value is: Among them, H is the fire evaluation value corresponding to the preset fire evaluation index, n is the total number of relevant monitoring data corresponding to the preset fire evaluation index at the current monitoring point, pi is the deviation coefficient of the i-th relevant monitoring data corresponding to the preset fire evaluation index, and ai is the weight coefficient of the i-th relevant monitoring data corresponding to the preset fire evaluation index.

5. The fire warning method for an electrochemical energy storage system according to claim 4, characterized in that: Based on the analysis results and the application coefficient of the corresponding preset fire evaluation index, a fire risk coefficient of the corresponding monitoring point is generated, including: Obtaining application information of each preset fire evaluation index in the historical fire log, wherein the application information includes timeliness information, application information, and accuracy information; Determine the discovery time node and judgment time node of the corresponding preset fire evaluation indicator according to the timeliness information of each preset fire evaluation indicator in each historical fire log, and compare them with the standard discovery time node and standard judgment time node in the corresponding historical fire log to obtain the first time deviation and the second time deviation of each preset fire evaluation indicator in each historical fire log, and generate the timeliness of the corresponding preset fire evaluation indicator according to the multiple first time deviations and the second time deviations; Determining the application form of the preset fire evaluation index in the corresponding historical fire log according to the application information of each preset fire evaluation index in the historical fire log, wherein the application form includes single application and joint application; When used alone, determine the number of times the corresponding preset fire evaluation index is used alone in all historical fire logs, and when used alone, the first influence of the corresponding preset fire evaluation index on the fire warning results of the historical fire logs; When it is a joint application, determining the number of joint applications of the corresponding preset fire evaluation index in all historical fire logs, and the second influence of the corresponding preset fire evaluation index on the fire warning results of the historical fire logs when the joint application is performed; Generate the applicability of corresponding preset fire assessment indicators according to the plurality of first impact degrees and second impact degrees; Determine the similarity between the historical fire evaluation value of the preset fire evaluation index and the fire warning result of each historical fire log according to the accurate information of each preset fire evaluation index in the historical fire log, and generate the accuracy of the corresponding preset fire evaluation index according to the number of historical fire logs with similarity greater than a preset similarity threshold and the number of all historical fire logs; Generate an application coefficient corresponding to each preset fire evaluation indicator according to the timeliness, operability and accuracy of each preset fire evaluation indicator; The calculation formula of the application coefficient is: Wherein, Y is the application coefficient, k1 is the first application conversion coefficient, c1 is the weight coefficient of timeliness, t1s is the first time deviation of the preset fire evaluation index in the sth historical fire log, b1 is the weight coefficient of the first time deviation, t2s is the second time deviation of the preset fire evaluation index in the sth historical fire log, b2 is the weight coefficient of the second time deviation, m1 is the total number of historical fire logs, k2 is the second application conversion coefficient, w1 is the weight coefficient of the first influence, m2 is the number of historical fire logs in which the preset fire evaluation index is used alone in the historical fire log, q1z is the first influence of the preset fire evaluation index in the zth historical fire log used alone, w2 is the weight coefficient of the second influence, q2v is the second influence of the preset fire evaluation index in the vth historical fire log used jointly, m3 is the number of historical fire logs in which the preset fire evaluation index is used jointly in the historical fire log, k3 is the third application conversion coefficient, and m4 is the number of historical fire logs whose similarity is greater than the preset similarity threshold; Generate a fire risk coefficient for the corresponding monitoring point based on the fire evaluation values ​​of all preset fire evaluation indicators of each monitoring point in the analysis results and the application coefficient of the corresponding preset fire evaluation indicators; The calculation formula of the fire risk coefficient of the monitoring point is: Among them, F is the fire risk coefficient, Hz is the fire evaluation value of the z-th preset fire evaluation index, and Yz is the application coefficient of the z-th preset fire evaluation index.

6. The fire warning method for an electrochemical energy storage system according to claim 5, characterized in that: Based on the fire risk coefficient of all monitoring points, determine whether there are abnormal monitoring points, including: Presetting a first preset fire risk coefficient and a second preset fire risk coefficient; When the fire risk coefficient of the monitoring point is less than the first preset fire risk coefficient, the corresponding monitoring point is judged to be a healthy monitoring point; When the fire risk coefficient of the monitoring point is between the first preset fire risk coefficient and the second preset fire risk coefficient, the corresponding monitoring point is judged to be a general monitoring point; When the fire risk coefficient of the monitoring point is greater than the second preset fire risk coefficient, the corresponding monitoring point is judged to be an abnormal monitoring point.

7. The fire warning method for an electrochemical energy storage system according to claim 6, characterized in that: Also includes: The general monitoring points are set as characteristic monitoring points, and the data change characteristics of the relevant monitoring data corresponding to the current characteristic monitoring points in the future period are predicted based on the historical fire data of the characteristic monitoring points; The judgment results of the current characteristic monitoring points are updated according to the data change characteristics of the relevant monitoring data in the future period.

8. The electrochemical energy storage system fire warning method according to claim 7, characterized in that: The correction coefficient of the fire risk coefficient of the current characteristic monitoring point is generated based on the predicted monitoring data of the characteristic monitoring point in the future period, including: Obtain historical fire data of historical characteristic periods of multiple historical fire logs at each characteristic monitoring point, take the characteristic period of each historical fire log as a time reference line, set historical monitoring time nodes according to preset time intervals, collect historical fire data based on each historical monitoring time node, and map it to the corresponding time reference line to obtain a fire data change curve for each historical fire log; Determine the fire change characteristics of the corresponding historical fire data according to the fire data change curve of each historical fire log, wherein the fire change characteristics include the historical data deviation value between the historical fire data and the standard monitoring data interval and the historical fluctuation value of the historical fire data at each historical monitoring time node; According to the monitoring data analysis diagram of the relevant monitoring data of each characteristic monitoring point, the relevant monitoring data whose fluctuation degree at the corresponding characteristic monitoring point is greater than the preset fluctuation degree is screened out, and the real-time characteristic time period of the screened relevant monitoring data and the change characteristics to be judged within the real-time characteristic time period are determined, wherein the change characteristics to be judged include the real-time data deviation value between the relevant monitoring data and the standard monitoring data interval and the real-time fluctuation value of the relevant monitoring data at each monitoring time node in the real-time characteristic time period, wherein the real-time characteristic time period is smaller than the historical characteristic time period; the change characteristics to be judged of the screened relevant monitoring data are respectively compared with multiple fire change characteristics of historical fire data of the same data type, and the similarity between the change characteristics to be judged and the fire change characteristics is generated according to the comparison results; If the similarity between the change feature to be judged and multiple fire change features is greater than the preset similarity threshold, the data change characteristics of the corresponding relevant monitoring data in the future time period are determined according to the fire change feature with the greatest similarity, and the judgment result of the characteristic monitoring point is updated to the abnormal monitoring point. If the similarities between the change feature to be judged and multiple fire change features are all less than the preset similarity threshold, the judgment result of the feature monitoring point is updated to a healthy monitoring point.

9. The electrochemical energy storage system fire warning method according to claim 8, characterized in that: If it exists, generate corresponding warning information and generate warning instructions, including: If there are abnormal monitoring points, determine the location information of the abnormal monitoring points and the abnormal related monitoring data, and combine the fire risk coefficient of the abnormal monitoring points to generate corresponding warning information and send warning instructions.

Images