Intelligent circuit breaker function adaptive adjustment method based on artificial intelligence

Abstract

The invention discloses an intelligent circuit breaker function adaptive adjustment method based on artificial intelligence, and relates to the technical field of circuit breaker function adjustment. The method comprises the steps of obtaining a circuit loop controlled by a circuit breaker, collecting user data and environment data of the circuit loop, and obtaining different types of fault probabilities of the circuit loop; searching a fault type corresponding to the circuit breaker function and recording the fault type as a functional fault type, and obtaining a fault probability corresponding to the circuit breaker function and recording the fault probability as a functional fault probability; collecting historical fault data, and calculating to obtain a functional fault degree; obtaining a demand distribution value of the circuit breaker function in combination with the function fault probability and the function fault degree; acquiring cooperative equipment of the circuit breaker function, acquiring cooperative data, and analyzing to obtain a cooperative capability value of the circuit breaker function; and obtaining a self-adaptive adjustment scheme according to the cooperative capability value and the demand allocation value. According to the invention, the adaptability of function adaptive adjustment and application scenes of the intelligent circuit breaker is improved.

Description

Technical Field

[0001] This application relates to the technical field of circuit breaker function adjustment, and particularly relates to an intelligent circuit breaker function adaptive adjustment method based on artificial intelligence. Background Art

[0002] A circuit breaker refers to a switching device that can close, carry, and interrupt the current under normal circuit conditions and can close, carry, and interrupt the current under abnormal circuit conditions within a specified time. Circuit breakers have many functions and have been widely used. Because circuit breakers have many functions, how to coordinate the functions of circuit breakers so that the circuit breakers can better adapt to the application environment and meet the requirements of all functions is a problem waiting to be solved. In the prior art, the adaptive adjustment of circuit breaker functions only adjusts according to the monitored current and voltage data, ignoring the requirements of the application scenario. Therefore, the adaptability of the intelligent circuit breaker function adaptive adjustment to the application scenario is not high, and the user experience is not good. Summary of the Invention

[0003] The purpose of the present invention is to provide an intelligent circuit breaker function adaptive adjustment method based on artificial intelligence to solve the problems raised in the above background art.

[0004] An intelligent circuit breaker function adaptive adjustment method based on artificial intelligence provided by this application adopts the following technical solutions: Obtain the circuit loop controlled by the circuit breaker, collect the user data and environmental data of the circuit loop, and obtain the failure probabilities of different types of the circuit loop according to the user data and environmental data; Obtain the circuit breaker functions, find the failure types corresponding to the circuit breaker functions and record them as function failure types, and superimpose the failure probabilities of all function failure types to obtain the failure probability corresponding to the circuit breaker function, which is recorded as the function failure probability; Collect historical failure data, and calculate the average failure degree of the function failure type according to the historical failure data as the function failure degree of the circuit breaker function; Set the proportionality factors of the function failure probability and the function failure degree respectively, and calculate the required resource allocation value of the circuit breaker function according to the proportionality factors; Obtain the collaborative devices of the circuit breaker function, collect the collaborative data of the collaborative devices, and analyze the collaborative ability value of the circuit breaker function according to the collaborative data; Adjust the resources according to the collaborative ability value and the required allocation value to obtain an adaptive adjustment plan, and the circuit breaker adjusts according to the adaptive adjustment plan.

[0005] Preferably, the step of obtaining the circuit loop controlled by the circuit breaker, collecting the user data and environmental data of the circuit loop, and obtaining the failure probabilities of different types of the circuit loop according to the user data and environmental data is specifically as follows: Obtain the environmental conditions under which different types of faults occur in the circuit loop, and extract the corresponding environmental parameter range from the environmental conditions, which is recorded as fault environmental data; Collect the real-time environmental data of the circuit loop, and compare to obtain the environmental similarity between the fault environmental data and the real-time environmental data; Establish a correlation curve between the environmental similarity and the fault probability, and find the fault probability corresponding to different fault types according to the correlation curve, which is recorded as the environmental fault probability; Collect the user data of the circuit loop, analyze the operation fault probability of different types of faults according to the user data, and superimpose the corresponding environmental fault probability to obtain the fault probability of different types of faults.

[0006] Preferably, the step of collecting the user data of the circuit loop and analyzing the operation fault probability of different types of faults according to the user data is specifically as follows: Obtain user data, where the user data includes user personal data and user operation data; Collect the operation standards corresponding to different types of faults in the circuit loop, and extract the reading information volume and reading frequency of the user reading the operation standards according to the user personal data; Extract the number of times the user accurately processes different types of faults in the circuit loop according to the user personal data, which is recorded as the accurate processing times; Combine the reading information volume, reading frequency and accurate processing times of the same fault, and comprehensively obtain the first fault probability; Analyze the second fault probability of different types of faults according to the user operation data, and superimpose the first fault probability of the corresponding type of fault to obtain the operation fault probability of different types of faults.

[0007] Preferably, the step of analyzing the second fault probability of different types of faults according to the user operation data is specifically as follows: Obtain user operation data, and judge whether the user operation is associated with the fault type according to the user operation data; If the user operation is associated with the fault type, extract the real-time operation steps from the user operation data; Obtain the historical operation data of the user, and extract the historical operation steps from the historical operation data; For the historical similarity between the obtained real-time operation steps and the historical operation steps, judge whether the historical similarity reaches a preset similarity threshold. If it reaches the preset similarity threshold, extract the fault probability of different types from the historical operation data as the second fault probability; If the preset similarity threshold is not reached, compare the standard similarity between the real-time operation steps and the standard operation steps, and obtain the standard difference degree according to the standard similarity, which is recorded as the second fault probability; If the user operation is not associated with the fault type, determine whether the historical similarity reaches a preset similarity threshold. If the preset similarity threshold is not reached, use the standard difference degree as the second fault probability; If the preset similarity threshold is reached, the second fault probability is 0.

[0008] Preferably, the step of obtaining the collaborative device of the circuit breaker function, collecting the collaborative data of the collaborative device, and obtaining the collaborative ability value of the circuit breaker function according to the collaborative data is specifically as follows: Obtain the collaborative device of the circuit breaker function, collect the collaborative data of the collaborative device, and determine whether the collaborative device has the ability to independently implement the circuit breaker function according to the collaborative data; If the collaborative device has the ability to independently implement the circuit breaker function, the collaborative ability value corresponding to the circuit breaker function is the maximum value; If the collaborative device does not have the ability to independently implement the circuit breaker function, determine whether the collaborative device has the monitoring ability; If the collaborative device has the monitoring ability, obtain the monitoring data of the collaborative device, and obtain the device ability value of the collaborative device according to the monitoring data; Select the maximum device ability value as the collaborative ability value of the circuit breaker function; If the collaborative device does not have the monitoring ability, obtain the communication data between the circuit breaker and the collaborative device, and obtain the collaborative ability value of the circuit breaker function according to the communication data.

[0009] Preferably, the step of if the collaborative device has the monitoring ability, obtain the monitoring data of the collaborative device, and obtain the device ability value of the collaborative device according to the monitoring data is specifically as follows: Obtain the historical transmission data of the collaborative device, and determine whether the collaborative device transmits data only during a fault; If the collaborative device transmits data only during a fault, obtain the historical monitoring data, and extract the fault monitoring accuracy rate; Extract the average transmission time of the data according to the historical transmission data, and combine the fault monitoring accuracy rate to obtain the device ability value of the collaborative device; If the collaborative device does not transmit data only during a fault, extract the data transmission interval time of the collaborative device; Establish a relationship table between the data transmission interval time and the timely response probability of the circuit breaker, and find the timely response probability of the circuit breaker according to the relationship table as the device ability value.

[0010] Preferably, the step of if the collaborative device does not have the monitoring ability, obtain the communication data between the circuit breaker and the collaborative device, and obtain the collaborative ability value of the circuit breaker function according to the communication data is specifically as follows: If the collaborative device does not have the monitoring ability, determine whether the collaborative device has the ability to control the circuit loop; If the collaborative device has the ability to control the circuit loop, obtain the communication data between the circuit breaker and the collaborative device; Extract the average reception duration of the collaborative device receiving the circuit breaker information from the communication data; Obtain the average occurrence duration of different faults from fault monitoring to fault occurrence in the functional fault types, and select the shortest average occurrence duration as the standard occurrence duration; Determine whether the average reception duration is not less than the standard occurrence duration. If the average reception duration is not less than the standard occurrence duration, obtain the average response duration of the collaborative device and record it as the device response duration; Collect the average response duration of the circuit breaker function and record it as the circuit breaker response duration, and calculate the ratio of the circuit breaker response duration to the device response duration as the collaboration ability value; If the collaborative device does not have the ability to control the circuit loop, obtain the function data of the collaborative device, and obtain the collaboration ability value according to the function data.

[0011] Preferably, the step of, if the collaborative device does not have the ability to control the circuit loop, obtaining the function data of the collaborative device and obtaining the collaboration ability value according to the function data is specifically as follows: Obtain the role played by the collaborative device in the realization of the circuit breaker function and record it as the auxiliary role; Statistically record the proportion of the number of circuit breakers using the auxiliary role to realize the function to the total number of circuit breakers as the usage proportion; Statistically record the proportion of the number of times the function realization fails due to the circuit breaker not using the auxiliary role to the total number of times the function is not realized using the auxiliary role as the failure proportion; Comprehensively obtain the collaboration ability value by combining the usage proportion and the failure proportion.

[0012] Preferably, the step of adjusting the resources according to the collaboration ability value and the demand allocation value to obtain an adaptive adjustment plan, and the circuit breaker adjusts according to the adaptive adjustment plan is specifically as follows: Normalize the collaboration ability values corresponding to different circuit breaker functions, and update the collaboration ability values of the circuit breaker functions; Establish a resource allocation equation set, obtain the solution of the resource allocation equation set, and obtain the determined allocation values of different circuit breaker functions; Form an adaptive adjustment plan according to the determined allocation values, and the circuit breaker adjusts according to the adaptive adjustment plan; Resource allocation equation set: Resource allocation equation set: Resource allocation equation set:

[0013] Among them, 、 …… is the collaborative ability value for different circuit breaker functions, 、 …… are the determined allocation values for different circuit breaker functions and are the unknowns of the system of equations, 、 …… are the demand allocation values for different circuit breaker functions. n is the number of different circuit breaker functions and is a positive integer. Solve the unknowns of the system of equations according to the equation.

[0014] In summary, the present application includes at least one of the following beneficial technical effects: 1. Estimate the first fault probability of a fault occurrence based on the reading information volume, reading frequency, and accurate processing times. Estimate the second fault probability of the occurrence of the fault method in different cases according to whether the user operation is related to the fault and whether the similarity between the user's historical operation steps and the real-time operation steps reaches a preset threshold. Combine the first fault probability, the second fault probability, and the environmental fault probability to obtain the fault probability. Superimpose the probabilities of all faults corresponding to the circuit breaker functions, and combine the fault degree to obtain the demand allocation value. By estimating the fault probability from multiple aspects, the function adaptive adjustment is further realized, and the accuracy of the function adaptive adjustment of the intelligent circuit breaker is improved.

[0015] 2. Through the assistance of the collaborative device in realizing the circuit breaker function, that is, the collaborative ability value, establish a system of equations in combination with the demand allocation value, and form an adaptive adjustment scheme through the solution of the system of equations to realize the adaptive adjustment of the circuit breaker function. Combine the collaborative ability of the collaborative device and the function requirements through the establishment of the system of equations to obtain the actual determined allocation value, and the result is more accurate. And obtaining the result through the system of equations is more convenient and comprehensive, improving the convenience and comprehensiveness of the function adaptive adjustment of the intelligent circuit breaker.

[0016] 3. Judge whether the collaborative device has the ability to independently realize the circuit breaker function, judge whether the collaborative device has the monitoring ability, and judge whether the collaborative device has the ability to control the circuit loop, and give different evaluation methods for the collaborative ability value of the collaborative device for different judgment results. Through multi-layer judgment, a more accurate usage situation can be obtained, so as to obtain a more accurate adaptive adjustment scheme, and the adaptive adjustment scheme is also more adapted to the usage scenario, improving the adaptability of the function adaptive adjustment of the intelligent circuit breaker to the application scenario. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a schematic diagram of the specific steps of an embodiment of a method for adaptively adjusting the function of an intelligent circuit breaker based on artificial intelligence according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] The following combines the embodiments and Figure 1A further detailed description of the present invention is provided, but the implementation manners of the present invention are not limited thereto.

[0019] The present invention discloses an artificial intelligence-based intelligent circuit breaker function adaptive adjustment method, which specifically includes the following steps: Step S1: Obtain the circuit loop controlled by the circuit breaker, collect the user data and environmental data of the circuit loop, and obtain the fault probabilities of different types of the circuit loop according to the user data and environmental data.

[0020] Step S2: Obtain the circuit breaker functions, find the fault types corresponding to the circuit breaker functions and record them as function fault types, and superimpose the fault probabilities of all function fault types to obtain the fault probability corresponding to the circuit breaker function, which is recorded as the function fault probability.

[0021] The circuit breaker has multiple functions, such as overload protection, short-circuit protection, power distribution, etc. Different response functions need to be taken under different faults. For example, thermal faults and mechanical faults may both cause short-circuit protection. Therefore, the fault types corresponding to the short-circuit protection function include thermal faults, mechanical faults, etc. By superimposing the fault probabilities of all function fault types, the probability of starting the corresponding function of the circuit breaker can be obtained, so it is recorded as the function fault probability. In step S1, the fault probabilities of different types have been obtained in the content. That is, the fault probabilities of all fault types have been obtained, and it is considered that a fault probability database has been formed. The multiple types of faults and the probabilities of multiple types of faults required in step S2 can be found from the existing fault probability database, and multiple function fault types and corresponding function fault probabilities can be obtained.

[0022] Step S3: Collect historical fault data, and calculate the average fault degree of the function fault type according to the historical fault data, which is used as the function fault degree of the circuit breaker function.

[0023] Collect historical fault data and calculate the average loss caused by the fault as the average fault degree. For example, the functional fault types include thermal faults and mechanical faults, that is, all fault types corresponding to a circuit breaker function. Calculate the average fault degree of all thermal faults and mechanical faults in the historical data as the functional fault degree of the corresponding circuit breaker function. In each fault, the user will evaluate the fault degree, obtain the fault degree corresponding to this fault and store it. The average fault degree can be calculated by taking the average of the fault degrees evaluated by the user each time based on loss cost, affected range, etc. in the historical fault data. The evaluation method of the fault degree can be set by the user himself. In each fault, the user makes a specific record of the fault and evaluates the fault degree, that is, the fault degree can be extracted from the historical fault data and then the average value is calculated to obtain the average fault degree. Since the fault involves multiple types, different types of fault data can be normalized to obtain the same data form. For example, all historical fault data are the fault degrees obtained through evaluation. After normalization, all fault degrees have the same data form and can be calculated and processed together. For example, find the maximum value of the fault degree of thermal faults from the historical fault data. Suppose the maximum value is 10, and the minimum value of the fault degree of thermal faults is 0, then the fault degree range of thermal faults is 0 - 10. Similarly, find the maximum value of the fault degree of mechanical faults recorded by the user from the historical fault data. Suppose the maximum value of mechanical faults is 100 and the minimum value of mechanical faults is 0, then the fault degree range of mechanical faults is 0 - 100. The fault degree of thermal faults is 2 and the fault degree of mechanical faults is 40. Scale them proportionally to the unified maximum range, that is, expand the fault degree of thermal faults by ten times, then the fault degree of thermal faults is 20. By taking the ratio of the fault degree to the maximum value of the range, the ranges of different fault degrees are unified to obtain unified fault degree data. For example, the fault types corresponding to the function of circuit breaker A are thermal faults and mechanical faults. Then the average fault degree of function A is calculated as follows: the fault degree value of the first thermal fault is 80, the fault degree of the second thermal fault is 60, and the fault degree value of the first mechanical fault is 70. Then the average fault degree is (80 + 60 + 70) / 3 = 70, and 70 is used as the functional fault degree of the circuit breaker function.

[0024] Step S4: Set the proportionality factors of the functional fault probability and the functional fault degree respectively, and calculate the required allocation value of the resources of the circuit breaker function according to the proportionality factors.

[0025] When the probability of functional failure of the circuit breaker is higher, the circuit breaker requires more attention resources, so the demand allocation value is higher. Similarly, the greater the degree of functional failure, the more attention resources are needed. The circuit breaker can adjust the working parameters to achieve the scheduling of attention resources, allocate more attention to functions that are easier to start and functions that need to be started more, so as to respond to faults more timely and accurately. The failure probability and failure degree can be quantified into risk values through a weighted model based on risk quantification, and the demand allocation value is obtained through weighted summation or multiplication. For example, the proportionality factors of the functional failure probability and the functional failure degree are set to 40% and 60% respectively, and the functional failure probability and the functional failure degree are 70% and 40 respectively, then the demand allocation value is 70%×40% + 40×60% = 24.28.

[0026] Step S5: Obtain the collaborative devices of the circuit breaker function, collect the collaborative data of the collaborative devices, and obtain the collaborative ability value of the circuit breaker function according to the analysis of the collaborative data.

[0027] Step S6: Adjust the resources according to the collaborative ability value and the demand allocation value to obtain an adaptive adjustment plan, and the circuit breaker adjusts according to the adaptive adjustment plan.

[0028] In actual application, since the conditions of the circuit loop protected by the circuit breaker are different in different usage scenarios, the probability and type of faults occurring are also different. Therefore, the resources of the circuit breaker are allocated according to the actual usage data of the usage scenario, so as to give more attention to faults that are more likely to occur and content that needs to be protected more, which is beneficial to timely respond to faults and reduce losses. For example, different protection characteristic curves are set inside the circuit breaker, and these curves define the operating time and operating current in different fault situations. For example, the overload protection characteristic curve usually has an inverse time limit characteristic, that is, the greater the current, the shorter the operating time. The short-circuit protection characteristic curve requires the circuit breaker to cut off the fault circuit within a very short time (such as milliseconds). When the circuit is likely to be overloaded but less likely to be short-circuited, the protection characteristic curve of the circuit breaker can be adjusted to make it pay more attention to the overload situation. Specifically, the setting value of the short-circuit protection can be increased, and the setting value of the overload protection can be decreased at the same time, so that the circuit breaker will act faster when detecting the overload current.

[0029] The steps of obtaining the circuit loop controlled by the circuit breaker, collecting the user data and environmental data of the circuit loop, and obtaining the fault probabilities of different types of the circuit loop according to the user data and environmental data are specifically as follows: Step S11: Obtain the environmental conditions for different types of faults to occur in the circuit loop, and extract the corresponding environmental parameter ranges from the environmental conditions as the fault environmental data.

[0030] The environment can affect the failure probability of a circuit loop. When the circuit loop operates in a non-standard environment, the probability of a fault occurring increases, and the environmental data corresponding to different faults varies. For example, thermal faults are more susceptible to temperature. Therefore, this type of fault will occur when the temperature exceeds a certain range, and the temperature outside the non-standard temperature range is the fault temperature. Mechanical faults are less affected by temperature. Therefore, the temperature range for mechanical faults is larger, and the fault temperature range in the fault environmental data will be smaller.

[0031] Step S12: Collect the real-time environmental data of the circuit loop, and compare to obtain the environmental similarity between the fault environmental data and the real-time environmental data.

[0032] The real-time environmental data includes parameter data of environmental factors such as temperature, humidity, and meteorology. The environmental similarity between the fault environmental data and the real-time environmental data is obtained by comparing through the existing cosine similarity model. The meteorology parameter data is set manually. For example, the meteorology value is set manually from 0 - 100, with a clear day being 100, a rainy day being 30, a snowy day being 10, etc., and it can be set by the user. Fault environmental data: When the historical fault occurred, the environmental parameters were wind speed 12m / s, temperature 5°C, and meteorology being overcast 60. The wind speed range in this area is 0 - 20m / s, the temperature range is 0 - 40°C, and the meteorology range value is 0 - 100. The corresponding parameter data is respectively compared with the maximum value of its range to obtain the normalized vector [0.6, 0.125, 0.6]. Real-time environmental data: The current wind speed is 10m / s, the temperature is 10°C, and the meteorology is cloudy 80, with the vector being [0.5, 0.25, 0.8]. According to the cosine similarity model, the calculated environmental similarity is 96.9%. Among them, the temperature range, wind speed range, and meteorology range are obtained from the maximum and minimum values of the corresponding environmental factors in the historical environmental data of this area. High wind speeds will increase the mechanical load on the circuit breaker and its support structure, which may cause equipment deformation or even fracture. Excessive wind speeds may cause equipment such as high-voltage circuit breakers and disconnectors with slender structures to deform or even break. Wind speed may also indirectly cause faults by affecting the heat dissipation and insulation performance of the circuit breaker. For example, in a high-wind-speed environment, the heat dissipation conditions on the surface of the circuit breaker may change, thereby affecting its internal temperature distribution and the arc extinction process. In addition, wind speed may also bring impurities such as dust and dirt, which adhere to the surface of the circuit breaker, reducing its insulation performance and increasing the risk of faults. When the wind speed is too high, the wire or the circuit breaker body may cause metal fatigue due to wind vibration.

[0033] Step S13: Establish a correlation curve between the environmental similarity and the fault probability, and find the fault probability corresponding to different fault types according to the correlation curve, which is recorded as the environmental fault probability.

[0034] When the environmental similarity is greater, it indicates a greater probability of a fault occurring. Therefore, the environmental fault probability is higher.

[0035] Step S14: Collect user data of the circuit loop. Based on the analysis of the user data, obtain the operation fault probabilities of different types of faults, and superimpose the corresponding environmental fault probabilities to obtain the fault probabilities of different fault types.

[0036] In actual application, in addition to the environment affecting the fault probability of the circuit loop, the user's usage will also affect the fault risk of the circuit loop. Estimating the fault probability by combining the usage environment of the circuit loop and the user's operation situation is conducive to obtaining more accurate data.

[0037] The steps of collecting user data of the circuit loop and obtaining the operation fault probabilities of different types of faults based on the analysis of the user data are specifically as follows: Step S141: Obtain user data, where the user data includes user personal data and user operation data.

[0038] Step S142: Collect the operation standards corresponding to different types of faults of the circuit loop, and extract the reading information volume and reading frequency of the user reading the operation standards based on the user personal data.

[0039] The calculation method widely used in the data platform can be adopted: The calculation of the information volume of the user reading the operation standards usually combines indicators such as page views (PV), stay duration, interaction behaviors (likes / comments), and source channels, and quantifies the real reading value through weighted comprehensive scoring. The reading frequency refers to the average number of times the user reads the content per day, that is, counts the number of times the user accesses and reads the content every day, and calculates its average value to obtain the reading frequency. The operation standards are displayed on electronic devices such as computers and mobile phones that can count parameters such as page views, stay duration, and interaction behaviors. For example, if the operation standard manual is 50 pages and the user browses 3 of them, then the user's page views are 3 pages. The stay duration of the user on the reading page of the operation standard manual is counted as 15 minutes, and the total number of likes, copying the content, and comments is 10 times, then the interaction behavior is recorded as 10 times. The weight ratios of page views, stay duration, and the number of interaction behaviors are set as 20%, 40%, and 40% respectively, then the user's reading information volume is 3×20% + 15×40% + 10×40% = 10.6.

[0040] Step S143: Based on the user personal data, extract the number of times the user accurately processes different types of faults of the circuit loop, which is recorded as the accurate processing times.

[0041] That is, extract the number of times the user accurately handles different types of faults in the circuit loop according to the user's fault handling records. In the management of circuit breakers, the staff need to record each fault handling. According to this record, compare whether each fault handling is completely consistent with the operation standard. If it is completely consistent, it is considered an accurate handling. For example, when fault A occurs, the operation standard requires first detecting whether the current is abnormal. If the current is abnormal, restart the circuit breaker; if the current is not abnormal, close the circuit breaker. In the user's record, fault A occurs, the current is detected to be abnormal, and the circuit breaker is closed. Then the operation meets the operation standard and is determined to be an accurate handling, and count the number of all accurate handlings.

[0042] Step S144, combine the reading information volume, reading frequency, and accurate handling times of the same fault to comprehensively obtain the first fault probability.

[0043] Respectively set the weight ratios of the reading information volume, reading frequency, and accurate handling times of a fault, and calculate the first fault probability according to the weight ratios. For example, the weight ratios of the reading information volume, reading frequency, and accurate handling times are respectively set to 60%, 20%, and 20%. The reading information volume, reading frequency, and accurate handling times are 0.5b, 0.2 times / day, and 1 respectively. Then the first fault probability is 0.5×60% + 0.2×20% + 1×20% = 64%.

[0044] Step S145, obtain the second fault probability of different types of faults according to the user operation data analysis, and superimpose the first fault probability of the corresponding type of fault to obtain the operation fault probability of different types of faults.

[0045] In actual application, the first fault probability inferred from the user's personal data is actually the fault probability obtained from the basic situation of the user's own operation. Due to individual differences, some users are more cautious during use, so the probability of faults will be greatly reduced, while some users are careless, so the probability of faults will be greatly increased. And the greater the reading information volume and reading frequency of the user, it means that the user pays more attention to safe operation and has a stronger safety awareness. And the more accurate handling times, it means that the user can better practice the operation standard, better handle faults, thereby reducing the occurrence of faults, having a stronger ability to respond to faults, and being able to more quickly detect fault signs, thereby reducing the fault probability or reducing the startup probability of the circuit breaker function.

[0046] The steps to obtain the second fault probability of different types of faults according to the user operation data analysis are specifically as follows: Step S1451, obtain the user operation data, and judge whether the user operation is related to the fault type according to the user operation data.

[0047] Some faults in the circuit loop are related to user operations, while some are not. For example, when users overload the device, it may cause an overload fault, but not an open circuit fault. Therefore, the overloading operation has no association with the open circuit fault. Based on historical fault data, the user operations corresponding to different fault causes are extracted to form an operation-fault table, and it is judged whether there is an association according to the operation-fault table. That is, if the corresponding user operation cause appears in the historical fault, it is judged as associated.

[0048] Step S1452, if the user operation is associated with the fault type, extract the real-time operation steps from the user operation data.

[0049] Step S1453, obtain the historical operation data of the user, and extract the historical operation steps from the historical operation data.

[0050] Step S1454, for the historical similarity between the obtained real-time operation steps and the historical operation steps, judge whether the historical similarity reaches the preset similarity threshold. If it reaches the preset similarity threshold, extract different types of fault probabilities from the historical operation data as the second fault probability.

[0051] If the user operation is related to the fault type, extract the historical operation steps, and compare the historical similarity between the real-time operation steps and the historical operation steps. The historical similarity can be obtained through existing technologies such as cosine similarity and Pearson correlation coefficient. By using cosine similarity and calculating according to the calculation method of environmental similarity, the historical similarity is obtained. If the historical similarity reaches the similarity threshold, it means that the user operation is normal and follows the past operations. Therefore, the fault probability in the historical operation data is used as the second fault probability. For example, if the probability of load occurrence in the historical operation is 30%, and the real-time operation follows the historical operation, the second fault probability is 30%.

[0052] Step S1455, if the preset similarity threshold is not reached, compare the standard similarity between the real-time operation steps and the standard operation steps, and obtain the standard difference degree according to the standard similarity and record it as the second fault probability.

[0053] The standard difference degree is obtained through 1 - standard similarity.

[0054] The standard similarity is obtained through the cosine similarity model. Suppose the standard similarity is 30%, then the difference degree is 1 - 30% = 70%.

[0055] If the preset similarity threshold is not reached, it indicates that the user's status is offline today. When the status is offline, the probability of errors will increase. Therefore, by comparing the real-time operation steps with the standard operation steps, the standard difference is recorded as the second probability. For example, the difference between the user's operation and the standard operation is 50%, but the probability of faults occurring in the user's historical operations is only 10%. This shows that the user is relatively clear about the boundaries of faults and can adjust the operation according to the actual situation to reduce the probability of faults. When the similarity between the user and the historical operation does not reach the threshold, it means that the user's status is also not good, and then the probability of faults is likely to increase, so the difference is used as the second fault probability.

[0056] In step S1456, if the user's operation has no association with the fault type, it is judged whether the historical similarity reaches the preset similarity threshold. If the preset similarity threshold is not reached, the standard difference is used as the second fault probability.

[0057] In step S1457, if the preset similarity threshold is reached, the second fault probability is 0.

[0058] In practical applications, if the user's operation has no association with the fault type, first judge whether the user's status is good according to the historical similarity. If the user's operation has no association with the fault type, but the user's status is offline, the user may perform misoperations associated with the fault type. Therefore, the standard difference is used as the second fault probability. If the user reaches the preset similarity threshold, indicating a good status, it is judged that the second fault probability is 0.

[0059] The steps of obtaining the collaborative device of the circuit breaker function, collecting the collaborative data of the collaborative device, and obtaining the collaborative ability value of the circuit breaker function according to the analysis of the collaborative data are specifically as follows: In step S51, obtain the collaborative device of the circuit breaker function, collect the collaborative data of the collaborative device, and judge whether the collaborative device has the ability to independently implement the circuit breaker function according to the collaborative data.

[0060] Sometimes the circuit breaker does not work alone and will work in collaboration with other devices to improve work efficiency. For example, in scenarios such as smart homes or smart hotels, the circuit breaker can work in collaboration with devices such as smart lighting, smart temperature control, and smart security. When the smart door lock recognizes that a guest has entered the room, the circuit breaker can automatically turn on devices such as lighting and air conditioners.

[0061] In step S52, if the collaborative device has the ability to independently implement the circuit breaker function, the collaborative ability value corresponding to the circuit breaker function is the maximum value.

[0062] Some collaborative devices can completely independently implement a certain function of the circuit breaker for functions such as emergency backup. When the circuit breaker fails to work, the function protection circuit can be activated immediately. Since the collaborative device can independently complete this function, for this function, even if the circuit breaker pays no attention to this function at all, the collaborative device can still activate the function and protect the circuit in a timely manner. Therefore, the collaborative ability value is the maximum value, and the maximum value refers to the maximum value among all historical collaborative ability values.

[0063] Step S53, if the collaborative device does not have the ability to independently implement the function of the circuit breaker, then determine whether the collaborative device has monitoring ability.

[0064] Judge whether it has monitoring ability through the instruction manual, historical data, etc. of the collaborative device.

[0065] Step S54, if the collaborative device has monitoring ability, then obtain the monitoring data of the collaborative device, and obtain the device ability value of the collaborative device according to the monitoring data.

[0066] Step S55, select the maximum device ability value as the collaborative ability value of the circuit breaker function.

[0067] Step S56, if the collaborative device does not have monitoring ability, then obtain the communication data between the circuit breaker and the collaborative device, and obtain the collaborative ability value of the circuit breaker function according to the communication data.

[0068] In practical applications, the assistance capabilities of different collaborative devices are different. Due to the different assistance capabilities of the collaborative devices, their collaborative capabilities for the circuit breaker function are also different. And sometimes there may be more than one collaborative device for the circuit breaker function. Therefore, select the maximum device ability value among the device ability values of all collaborative devices as the collaborative ability value.

[0069] If the collaborative device has monitoring ability, then the steps of obtaining the monitoring data of the collaborative device and obtaining the device ability value of the collaborative device according to the monitoring data are specifically as follows: Step S541, obtain the historical transmission data of the collaborative device, and judge whether the collaborative device transmits data only during faults.

[0070] Step S542, if the collaborative device transmits data only during faults, then obtain the historical monitoring data and extract the fault monitoring accuracy rate.

[0071] Step S543, extract the average transmission time of the data according to the historical transmission data, and combine the fault monitoring accuracy rate to obtain the device ability value of the collaborative device.

[0072] Set the weight ratios of the average transmission time and the fault monitoring accuracy rate respectively, and calculate the device capability value according to the weight ratios. For example, if the weight ratios of the average transmission time and the fault monitoring accuracy rate are set to 40% and 60% respectively, and the average transmission time and the fault monitoring accuracy rate are 1 second and 80% respectively, then the device capability value is 1×40% + 60%×80% = 0.88.

[0073] Step S544: If the collaborative device does not only transmit data during a fault, extract the data transmission interval time of the collaborative device.

[0074] Step S545: Establish a relationship table between the data transmission interval time and the timely response probability of the circuit breaker, and find the timely response probability of the circuit breaker according to the relationship table as the device capability value.

[0075] In practical applications, if the collaborative device does not have the ability to independently complete the circuit breaker function, then the collaborative device may have the ability to monitor data for the circuit breaker. Therefore, if the collaborative ability has monitoring ability, obtain the monitoring data of the collaborative device. After the collaborative device monitors the data, it will be transmitted to the circuit breaker to assist the circuit breaker in completing the work, and the monitoring and transmission methods of different monitoring devices are different. Some collaborative devices only have a monitoring function, so the monitored data is directly transmitted to the circuit breaker. And the longer the data transmission interval time, the more difficult it is to transmit the fault data to the circuit breaker in a timely manner. Therefore, the timely response probability of the circuit breaker is smaller, so the device capability value will be lower. For example, the collaborative device transmits data once every minute, and the circuit loop has failed at 38 seconds, but the data is not transmitted until 1 minute. In this way, the circuit breaker can respond at least 22 seconds later, so the device capability value is small. And if the collaborative device has an analysis ability, it can send data to the circuit breaker when a fault is detected, and the circuit breaker can respond in a timely manner to handle the fault. However, the analysis of the collaborative device may not be accurate. Therefore, the higher the fault monitoring accuracy rate of the collaborative device and the shorter the data transmission time, the greater the device capability value and the more helpful it is to the circuit breaker.

[0076] If the collaborative device does not have the monitoring ability, the steps to obtain the collaborative ability value of the circuit breaker function according to the communication data between the circuit breaker and the collaborative device are as follows: Step S561: If the collaborative device does not have the monitoring ability, determine whether the collaborative device has the ability to control the circuit loop.

[0077] Step S562: If the collaborative device has the ability to control the circuit loop, obtain the communication data between the circuit breaker and the collaborative device.

[0078] Step S563: Extract the average reception duration for the collaborative device to receive the circuit breaker information according to the communication data.

[0079] Step S564: Obtain the average occurrence duration from fault monitoring to fault occurrence for different faults in the functional fault type, and select the shortest average occurrence duration as the standard occurrence duration.

[0080] The occurrence times of different faults also vary. For example, thermal faults are usually caused by overheating of the device, such as poor contact, excessive load, etc. The occurrence duration of such faults depends on the speed and degree of temperature rise. Generally, the time from detecting the overheat signal to the actual occurrence of the fault may be relatively long. Electrical faults often involve abnormalities in current and voltage, such as overload, short circuit, etc. Since the propagation speed of electrical signals is very fast, the time from detecting the fault signal to the actual occurrence of the fault is often very short, possibly at the millisecond level.

[0081] Step S565: Determine whether the average reception duration is not less than the standard occurrence duration. If the average reception duration is not less than the standard occurrence duration, obtain the average response duration of the collaborative device and record it as the device response duration.

[0082] Step S566: Collect the average response duration of the circuit breaker function and record it as the circuit breaker response duration, and calculate the ratio of the circuit breaker response duration to the device response duration as the collaboration ability value.

[0083] Step S567: If the collaborative device does not have the ability to control the circuit loop, obtain the role data of the collaborative device, and obtain the collaboration ability value according to the role data.

[0084] In practical applications, if the collaborative device does not have the monitoring ability, it may have the control ability, that is, receive information from the circuit breaker and control the circuit loop to assist the circuit breaker in realizing its function. If the average reception duration is less than the standard occurrence duration, the fault has already occurred and the collaborative device only receives the data, so the collaborative device cannot respond to the fault in time, and thus the collaboration ability value is the minimum value. If the average reception duration is not less than the standard occurrence duration, the collaborative device can respond before the fault occurs. If the device response duration is shorter, the ratio of the circuit breaker response duration to the device response duration is larger, that is, for this function of the circuit breaker, the collaboration ability value is larger.

[0085] The steps of obtaining the role data of the collaborative device and obtaining the collaboration ability value according to the role data when the collaborative device does not have the ability to control the circuit loop are specifically as follows: Step S5671: Obtain the role played by the collaborative device in the realization of the circuit breaker function and record it as the auxiliary role.

[0086] If the collaborative device is neither used for monitoring data nor for control, then the role played by the collaborative device is generally auxiliary roles such as data transmission and access.

[0087] Step S5672: Count the proportion of the number of circuit breakers that use the auxiliary function to achieve the function to the total number of circuit breakers, and record it as the usage proportion.

[0088] Step S5673: Count the proportion of the number of times the function implementation fails due to the circuit breaker not using the auxiliary function to the total number of times the function is not implemented using the auxiliary function, and record it as the failure proportion.

[0089] Step S5674: Combine the usage proportion and the failure proportion to comprehensively obtain the collaborative ability value.

[0090] In actual application, the weight ratios of the usage proportion and the failure proportion are set respectively. Multiply the usage proportion and the failure proportion by their corresponding weight ratios respectively, and then add the calculation results after multiplication to obtain the collaborative ability value. When the usage proportion is higher, it indicates that the circuit breaker has a higher demand for this auxiliary function. And when the failure proportion is higher, it also indicates that this collaborative device can better help the circuit breaker achieve the function. Therefore, the comprehensive collaborative ability value is higher.

[0091] Adjust the resources according to the collaborative ability value and the demand allocation value to obtain an adaptive adjustment plan. The steps for the circuit breaker to adjust according to the adaptive adjustment plan are as follows: Step S61: Normalize the collaborative ability values corresponding to different circuit breaker functions to update the collaborative ability values of the circuit breaker functions.

[0092] Because the calculation methods of the collaborative ability values in different situations are different, it is difficult to compare and apply the calculation results of the collaborative ability values in different situations. Through normalization processing, the collaborative ability values can be better compared. For example, the collaborative ability value of A is 20, and its corresponding collaborative ability value range is 1 - 10. The collaborative ability value of B is 1, and its corresponding collaborative ability value range is 1 - 10. Then amplify the collaborative ability value of B to 10, so that the collaborative ability of A and the collaborative ability of B are comparable.

[0093] Step S62: Establish a resource allocation equation set, obtain the solution of the resource allocation equation set, and obtain the determined allocation values for different circuit breaker functions.

[0094] Step S63: Form an adaptive adjustment plan according to the determined allocation values, and the circuit breaker adjusts according to the adaptive adjustment plan.

[0095] Step S64: Resource allocation equation set:

[0096] Among them, 、 …… are the collaborative ability values of different circuit breaker functions, 、 …… Assign values to the determination of different circuit breaker functions and be the unknowns of the system of equations, 、 …… Assign values to the requirements of different circuit breaker functions. n is the number of different circuit breaker functions and is a positive integer. Solve the unknowns of the system of equations according to the equations.

[0097] In actual application, resources should be allocated according to the requirements of the circuit breaker functions. However, due to the different assistance capabilities of the cooperative devices for different functions, if the resources are allocated according to the original requirement assignment values, the resource allocation will be uneven due to the intervention of the cooperative devices. For example, the requirement assignment values for functions A, B, and C are 50, 30, and 20 respectively. After the resources are allocated according to the requirement assignment values, since the cooperative device for function A has a stronger ability, while the cooperative devices for functions B and C have weaker abilities, it results in an excess of resources for function A and insufficient resources for functions B and C. The ratio formed according to the requirement assignment values is 5:3:2, and the corresponding cooperative ability values are 70, 20, and 40 respectively. Then, substituting into the formula for calculation, the resource allocation percentages corresponding to functions A, B, and C are approximately 41.38%, 48.28%, and 10.34% respectively. It can be seen that since the cooperative ability of function A is stronger, the circuit breaker can allocate fewer resources to function A because the cooperative device will also respond in a timely manner to the faults corresponding to this function.

[0098] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereby. Therefore, all equivalent changes made according to the structure, shape, and principle of the present application shall be covered within the protection scope of the present application.

Claims

1. An intelligent circuit breaker function adaptive adjustment method based on artificial intelligence, characterized in that, Including the following steps: Obtain the circuit loop controlled by the circuit breaker, collect the user data and environmental data of the circuit loop, and obtain the failure probabilities of different types of the circuit loop according to the user data and environmental data; Obtain the functions of the circuit breaker, find the failure types corresponding to the functions of the circuit breaker and record them as functional failure types, superimpose the failure probabilities of all functional failure types, and obtain the failure probability corresponding to the functions of the circuit breaker, which is recorded as the functional failure probability; Collect historical failure data, and calculate the average failure degree of the functional failure type according to the historical failure data, which is used as the functional failure degree of the circuit breaker function; Set the proportionality factors of the functional failure probability and the functional failure degree respectively, and calculate the demand allocation value of the resources of the circuit breaker function according to the proportionality factors; Obtain the collaborative devices of the circuit breaker function, collect the collaborative data of the collaborative devices, and obtain the collaborative ability value of the circuit breaker function according to the analysis of the collaborative data; Adjust the resources according to the collaborative ability value and the demand allocation value to obtain an adaptive adjustment plan, and the circuit breaker adjusts according to the adaptive adjustment plan.

2. The method for adaptively adjusting the functions of an intelligent circuit breaker based on artificial intelligence according to claim 1, wherein The step of obtaining the circuit loop controlled by the circuit breaker, collecting the user data and environmental data of the circuit loop, and obtaining the failure probabilities of different types of the circuit loop according to the user data and environmental data is specifically as follows: Obtain the environmental conditions for the occurrence of different types of failures in the circuit loop, and extract the corresponding environmental parameter ranges from the environmental conditions and record them as failure environmental data; Collect the real-time environmental data of the circuit loop, and compare to obtain the environmental similarity between the failure environmental data and the real-time environmental data; Establish an association curve between the environmental similarity and the failure probability, and find the failure probabilities corresponding to different failure types according to the association curve, which are recorded as environmental failure probabilities; Collect the user data of the circuit loop, analyze the operation failure probabilities of different types of failures according to the user data, and superimpose the corresponding environmental failure probabilities to obtain the failure probabilities of different types of failures.

3. An intelligent circuit breaker function adaptive adjustment method based on artificial intelligence according to claim 2, characterized in that, The step of collecting the user data of the circuit loop and analyzing the operation failure probabilities of different types of failures according to the user data is specifically as follows: Obtain user data, where the user data includes user personal data and user operation data; Collect the operation standards corresponding to different types of failures in the circuit loop, and extract the reading information amount and reading frequency of the user reading the operation standards according to the user personal data; Extract the number of times the user accurately processes different types of failures in the circuit loop according to the user personal data, which is recorded as the accurate processing times; Combine the reading information amount, reading frequency and accurate processing times of the same failure, and comprehensively obtain the first failure probability; Analyze the second failure probabilities of different types of failures according to the user operation data, and superimpose the first failure probabilities of the corresponding types of failures to obtain the operation failure probabilities of different types of failures.

4. The intelligent circuit breaker function adaptive adjustment method based on artificial intelligence according to claim 3, characterized in that The step of analyzing the second failure probabilities of different types of failures according to the user operation data is specifically as follows: Obtain user operation data, and judge whether the user operation is associated with the failure type according to the user operation data; If the user operation is associated with the failure type, extract the real-time operation steps from the user operation data; Obtain the historical operation data of the user, and extract the historical operation steps from the historical operation data; For obtaining the historical similarity between the real-time operation steps and the historical operation steps, determining whether the historical similarity reaches a preset similarity threshold. If it reaches the preset similarity threshold, extract the failure probabilities of different types from the historical operation data as the second failure probability; If it does not reach the preset similarity threshold, compare the standard similarity between the real-time operation steps and the standard operation steps, and obtain the standard difference degree according to the standard similarity and record it as the second failure probability; If the user operation has no association with the failure type, determine whether the historical similarity reaches the preset similarity threshold. If it does not reach the preset similarity threshold, use the standard difference degree as the second failure probability; If it reaches the preset similarity threshold, the second failure probability is 0.

5. The intelligent circuit breaker function adaptive adjustment method based on artificial intelligence according to claim 1, characterized in that The steps of obtaining the collaborative device of the circuit breaker function, collecting the collaborative data of the collaborative device, and obtaining the collaborative ability value of the circuit breaker function according to the analysis of the collaborative data are specifically as follows: Obtain the collaborative device of the circuit breaker function, collect the collaborative data of the collaborative device, and determine whether the collaborative device has the ability to independently implement the circuit breaker function according to the collaborative data; If the collaborative device has the ability to independently implement the circuit breaker function, the collaborative ability value corresponding to the circuit breaker function is the maximum value; If the collaborative device does not have the ability to independently implement the circuit breaker function, determine whether the collaborative device has the monitoring ability; If the collaborative device has the monitoring ability, obtain the monitoring data of the collaborative device, and obtain the device ability value of the collaborative device according to the monitoring data; Select the maximum device ability value as the collaborative ability value of the circuit breaker function; If the collaborative device does not have the monitoring ability, obtain the communication data between the circuit breaker and the collaborative device, and obtain the collaborative ability value of the circuit breaker function according to the communication data.

6. The intelligent circuit breaker function adaptive adjustment method based on artificial intelligence according to claim 5, wherein The steps of if the collaborative device has the monitoring ability, obtain the monitoring data of the collaborative device, and obtain the device ability value of the collaborative device according to the monitoring data are specifically as follows: Obtain the historical transmission data of the collaborative device, and determine whether the collaborative device transmits data only during a failure; If the collaborative device transmits data only during a failure, obtain the historical monitoring data, and extract the failure monitoring accuracy rate; Extract the average transmission time of the data according to the historical transmission data, and combine the failure monitoring accuracy rate to obtain the device ability value of the collaborative device; If the collaborative device does not transmit data only during a failure, extract the data transmission interval time of the collaborative device; Establish a relationship table between the data transmission interval time and the circuit breaker's timely response probability, and find the circuit breaker's timely response probability according to the relationship table as the device ability value.

7. The method for adaptively adjusting the function of an intelligent circuit breaker based on artificial intelligence according to claim 5, characterized in that, The steps of if the collaborative device does not have the monitoring ability, obtain the communication data between the circuit breaker and the collaborative device, and obtain the collaborative ability value of the circuit breaker function according to the communication data are specifically as follows: If the collaborative device does not have the monitoring ability, determine whether the collaborative device has the ability to control the circuit loop; If the collaborative device has the ability to control the circuit loop, obtain the communication data between the circuit breaker and the collaborative device; Extract the average reception duration for the collaborative device to receive the circuit breaker information according to the communication data; Obtain the average occurrence duration from fault monitoring to fault occurrence for different faults in the function fault type, and select the shortest average occurrence duration as the standard occurrence duration; Judge whether the average reception duration is not less than the standard occurrence duration. If the average reception duration is not less than the standard occurrence duration, obtain the average response duration of the collaborative device and record it as the device response duration; Collect the average response duration of the circuit breaker function and record it as the circuit breaker response duration, and calculate the ratio of the circuit breaker response duration to the device response duration as the collaboration ability value; If the collaborative device does not have the ability to control the circuit loop, obtain the role data of the collaborative device, and obtain the collaboration ability value according to the role data.

8. A method for adaptively adjusting the functions of an intelligent circuit breaker based on artificial intelligence according to claim 7, characterized in that The step of obtaining the role data of the collaborative device and obtaining the collaboration ability value according to the role data if the collaborative device does not have the ability to control the circuit loop is specifically as follows: Obtain the role played by the collaborative device in the implementation of the circuit breaker function and record it as the auxiliary role; Count the proportion of the number of circuit breakers using the auxiliary role to achieve the function to the total number of circuit breakers and record it as the usage proportion; Count the proportion of the number of times the function implementation fails due to the circuit breaker not using the auxiliary role to the total number of times the function is not implemented using the auxiliary role and record it as the failure proportion; Comprehensively obtain the collaboration ability value by combining the usage proportion and the failure proportion.

9. A method for adaptively adjusting the functions of an intelligent circuit breaker based on artificial intelligence according to claim 1, characterized in that, The step of adjusting the resources according to the collaboration ability value and the demand allocation value to obtain an adaptive adjustment plan, and the circuit breaker adjusts according to the adaptive adjustment plan is specifically as follows: Normalize the collaboration ability values corresponding to different circuit breaker functions and update the collaboration ability values of the circuit breaker functions; Establish a resource allocation equation set, obtain the solution of the resource allocation equation set, and obtain the determined allocation values for different circuit breaker functions; Form an adaptive adjustment plan according to the determined allocation values, and the circuit breaker adjusts according to the adaptive adjustment plan; Resource Allocation Equation Set: Wherein, , …… are the collaborative ability values of different circuit breaker functions, , …… are the determined allocation values of different circuit breaker functions and are the unknowns of the equation set, , …… are the required allocation values of different circuit breaker functions. n is the number of different circuit breaker functions and is a positive integer. Solve the unknowns of the equation set according to the equation.

Images