Electric power safety communication intelligent early warning system and method based on big data

Abstract

The invention discloses an electric power safety communication intelligent early warning system and method based on big data, and relates to the technical field of intelligent early warning. According to the invention, through the analysis module, the occurrence time of each historical fault phenomenon is analyzed, the first occurrence time and the last occurrence time of each fault phenomenon are obtained, and the fault time prediction unit determines the occurrence time of each fault phenomenon according to the obtained first occurrence time and the last occurrence time of each fault phenomenon. The method comprises the following steps: acquiring big data, calculating the time for starting fault prediction, if the current time is the time for starting fault prediction, predicting a fault reason causing a fault phenomenon according to an acquired big data condition, carrying out early warning prompt on a prediction result, and automatically adjusting the fault reason. Guarantee is provided for safe operation of communication equipment; meanwhile, an early warning prompt function is set for preventing faults, and the concept of intelligent early warning is met.

Description

Technical Field

[0001] The present invention relates to the field of intelligent early warning technology, and in particular to an intelligent early warning system and method for power safety communication based on big data. Background Art

[0002] Substations are an indispensable and important part of the power system. High-voltage transmission lines cross mountains, rivers, lakes and seas, and efficiently transmit the electricity generated by large power stations far away to various substations. In substations, communication equipment plays a vital role in ensuring information exchange and collaborative work between power equipment such as transformers, mutual inductors, and circuit breakers. Workers use communication equipment to remotely monitor and ensure the normal operation of various power equipment, gradually reduce the voltage, and then transmit it to thousands of households and various industrial users through the distribution network.

[0003] Therefore, once the communication equipment fails, it may cause operators to make mistakes when receiving or sending instructions, increasing the risk of misoperation, causing power equipment damage, power outages and other accidents, or failing to communicate with the dispatch center and other emergency agencies in a timely manner, delaying emergency response time and causing accidents. At present, the fault detection of communication equipment mainly relies on regular inspections by staff, which lacks real-time performance; at the same time, for some potential fault problems, staff cannot accurately identify them, and omissions occur, resulting in communication equipment failures. Therefore, there is an urgent need for an intelligent early warning system and method for power safety communication based on big data to solve the above problems. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent early warning system and method for power safety communication based on big data to solve the problems raised in the prior art.

[0005] To achieve the above object, the present invention provides the following technical solution: an intelligent early warning method for power safety communication based on big data, the early warning method comprising:

[0006] Step S1: Collect each historical fault record in the communication device, wherein each historical fault record includes the fault phenomenon, occurrence time, and fault cause;

[0007] Each of the collected historical fault records only contains one fault phenomenon and one fault cause that causes the fault phenomenon.

[0008] Step S2: Analyze the occurrence time of the same fault phenomenon in all historical fault records to obtain the first occurrence time and the last occurrence time of the same fault phenomenon;

[0009] Step S3: Calculate the fault phenomenon that will occur and predict the time when the fault phenomenon starts according to the first occurrence time and the last occurrence time of the fault phenomenon acquired;

[0010] Step S4: If the current time is the time when the fault phenomenon starts to be predicted, the system calculates the probability of each fault cause causing the fault phenomenon to occur;

[0011] Step S5: The system issues an early warning prompt based on the probability of occurrence of each fault cause that leads to the fault phenomenon.

[0012] Furthermore, the process of obtaining the time when the fault phenomenon first occurs and the time when the fault phenomenon last occurs in step S2 includes:

[0013] Step S2-1: Perform time series analysis on the occurrence time of the same fault phenomenon in all historical fault records to obtain the first occurrence time Tmin and the last occurrence time Tmax of the fault phenomenon;

[0014] Step S2-2: The system integrates the acquired Tmin and Tmax to obtain a set {(χ1, Tmin1, Tmax1), ..., (χ i , Tmin i , Tmax i ), …};

[0015] Among them, χ i Indicates the i-th fault phenomenon;

[0016] In the above steps, the occurrence of the fault phenomenon is considered to have a certain regularity. In the present invention, the time regularity of the occurrence of the fault phenomenon is mainly analyzed. The fault phenomenon may occur in a specific period of time. Therefore, by obtaining the time when the fault phenomenon first occurs and the time when the fault phenomenon last occurs, it represents that the fault phenomenon occurs within a period of time, and prepares for subsequent fault prediction.

[0017] Furthermore, the process of calculating the time threshold of the fault phenomenon that will occur and starting to predict the fault phenomenon that will occur in step S3 includes:

[0018] Step S3-1: The system obtains the current time T, and compares T with the set {(χ1, Tmin1, Tmax1), ..., (χ i , Tmin i , Tmax i ), …}, and extract the Tmin greater than T, and obtain the set {Tmin1, Tmin2, …, Tmin i}, compare the difference between Tmin and T in the set, and the Tmin with the smallest difference m Corresponding fault phenomenon χ m is the fault phenomenon to be predicted;

[0019] Step S3-2: Compare T with the set {(χ1, Tmin1, Tmax1), ..., (χ i , Tmin i , Tmax i ), …}, and extract the Tmax smaller than T to obtain the set {Tmax1, Tmax2, …, Tmax m}, compare the difference between Tmax and T in the set, and the Tmax with the smallest difference k is the fault phenomenon predicted last time k time of last appearance;

[0020] Step S3-3: Obtaining Tmax k With Tmin m At the middle moment, according to the formula:

[0021]

[0022] Wherein, ΔT represents the time threshold for starting fault prediction for the fault phenomenon that is about to occur;

[0023] The above steps are for calculating when the system starts to predict the fault of the communication equipment. Because fault prediction takes time, the fault should be predicted before the above calculated fault occurrence time period. Moreover, the premise for the next fault prediction should be the end of the previous fault prediction, that is, the last time the previous fault phenomenon occurred. Therefore, the time of fault prediction is the middle moment between the last time the previous fault phenomenon occurred and the first time the next fault phenomenon occurs.

[0024] Furthermore, in step S4, the process in which the system calculates the probability of occurrence of each fault cause causing the fault phenomenon includes:

[0025] Step S4-1: Based on the historical fault phenomenon X = {χ1, χ2, ..., χ n} and historical fault causes M={m1,m2,…,m n}, extract all historical fault phenomena corresponding to the same fault cause, and form a set χ={χ1,χ2,…,χ k}, where χ is a subset of the set X;

[0026] Step S4-2: Calculate the fault cause m under the condition that the same fault phenomenon χσ occurs.i The probability of occurrence is based on the formula:

[0027]

[0028] in,

[0029] Among them, P(χσ|m i ) is the fault cause m i The probability of the fault phenomenon χσ occurring under the condition of occurrence; N χ The fault reason is m i The number of records of the corresponding fault phenomenon χ in X; N χσ is the number of records of fault phenomenon χσ in set χ; P(m i ) is among all the fault causes m i Probability of occurrence; N M is the number of records of all fault causes in set M; N mi The fault reason is m i The number of records in the set M; P(χσ) is the number of records in all fault phenomena σ Probability of occurrence; N X is the number of records of all fault phenomena in X; N Xσ The fault phenomenon is σ The number of records in set X;

[0030] In the above steps, one fault phenomenon may correspond to multiple fault causes, and one fault cause may also lead to the occurrence of multiple fault phenomena. When a certain fault phenomenon is predicted to occur, the probability of all fault causes occurring under the condition of the occurrence of the fault phenomenon is calculated to prepare for the subsequent screening of the fault causes that cause the fault phenomenon.

[0031] Furthermore, the process of the system performing an early warning prompt in step S5 includes:

[0032] Step S5-1: The system obtains the probability of each fault cause causing the fault phenomenon, and selects the fault cause with the highest probability through a comparator;

[0033] Step S5-2: The system detects the parameters of the communication device according to the cause of the fault. When it is identified that the parameters exceed the threshold range, a first-level warning is triggered, and the parameters are automatically adjusted. If they return to the threshold range, the first-level warning disappears;

[0034] Step S5-3: If the parameter cannot be restored to within the threshold range, the system issues a secondary warning and sends the cause of the fault to the application end connected to the communication device;

[0035] The multi-level warning mechanism is set in the above steps to prevent the situation where the fault cannot be automatically repaired. In this case, the second-level warning is used to remind the staff to repair the communication equipment.

[0036] In order to better implement the above method, an intelligent early warning system for power safety communication based on big data is also proposed. The early warning system includes an acquisition module, a transmission module, an analysis module, a fault prediction module, and an early warning module;

[0037] A collection module, used to collect each historical fault record in the communication device, wherein each fault record includes the fault phenomenon, occurrence time, and fault cause;

[0038] Transmission module, used for transmission of various information between modules;

[0039] An analysis module, which analyzes the occurrence time of the same fault phenomenon in all historical fault records to obtain the first and last occurrence time of the same fault phenomenon, and calculates the time to start predicting the fault phenomenon based on the first and last occurrence time of the fault phenomenon;

[0040] The fault prediction module is used to calculate the fault phenomenon that will occur and calculate the probability of each fault cause that causes the fault phenomenon to occur;

[0041] The early warning module is used to screen out the causes of the faults, detect the parameters of the communication equipment according to the causes of the faults, and give early warning prompts.

[0042] Further, the analysis module includes a fault occurrence time analysis unit and a fault prediction time analysis unit;

[0043] A fault occurrence time analysis unit, used to sort the same fault phenomenon in all historical fault records according to the occurrence time, and obtain the first occurrence time and the last occurrence time of the same fault phenomenon;

[0044] The fault prediction time analysis unit is used to calculate the time to start predicting the fault phenomenon according to the acquired time of the first occurrence and the last occurrence of the fault phenomenon.

[0045] Further, the fault prediction module includes a fault phenomenon prediction unit and a fault cause prediction unit;

[0046] A fault phenomenon prediction unit is used to analyze the time when the fault phenomenon first occurs and obtain the fault phenomenon to be predicted;

[0047] The fault cause prediction unit calculates the probability of a certain fault cause occurring when the same fault phenomenon occurs, based on the acquired number of historical fault phenomenon records and the number of historical fault cause records.

[0048] Furthermore, the warning module includes a primary warning unit and a secondary warning unit;

[0049] The first-level warning unit is used to screen out the cause of the fault and detect the parameters of the communication equipment according to the cause of the fault. If the parameter exceeds the threshold range, the first-level warning is triggered;

[0050] The secondary warning unit is used to identify parameters that cannot be restored to the threshold range and trigger the secondary warning.

[0051] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention analyzes the time of occurrence of each historical fault phenomenon through an analysis module, obtains the time of the first occurrence and the last occurrence of each fault phenomenon, and the fault time prediction unit calculates the time to start fault prediction by obtaining the time of the first occurrence and the last occurrence of each fault phenomenon. If the current time is the time to start fault prediction, the cause of the fault that caused the fault phenomenon is predicted based on the acquired big data, the predicted result is warned and the cause of the fault is automatically adjusted, thereby providing a guarantee for the safe operation of the communication equipment; at the same time, the early warning prompt function provided for preventing faults is in line with the concept of intelligent early warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 A schematic diagram of a flow chart of an intelligent early warning method for power safety communication based on big data according to the present invention;

[0053] Figure 2 This is a structural schematic diagram of an electric power safety communication intelligent early warning system based on big data in the present invention. DETAILED DESCRIPTION

[0054] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0055] Example: Figure 1-Figure 2 As shown, the present invention provides a technical solution, an intelligent early warning method for power safety communication based on big data, the early warning method comprising:

[0056] Step S1: Collect each historical fault record in the communication device, wherein each historical fault record includes the fault phenomenon, occurrence time, and fault cause;

[0057] In an embodiment of the present invention, a staff member installs the application end of the present invention on a computer or a mobile phone, and the application end monitors the communication equipment in real time. When a communication equipment fails, the application end uploads the fault record to the cloud server through the computer or mobile phone, and the cloud server stores the fault record.

[0058] Step S2: Analyze the occurrence time of the same fault phenomenon in all historical fault records to obtain the first occurrence time and the last occurrence time of the same fault phenomenon;

[0059] The process of obtaining the time when the fault phenomenon first occurs and the time when the fault phenomenon last occurs in step S2 includes:

[0060] Step S2-1: Perform time series analysis on the occurrence time of the same fault phenomenon in all historical fault records to obtain the first occurrence time Tmin and the last occurrence time Tmax of the fault phenomenon;

[0061] Step S2-2: The system integrates the acquired Tmin and Tmax to obtain a set {(χ1, Tmin1, Tmax1), ..., (χ i , Tmin i , Tmax i ), …};

[0062] Among them, χ i Indicates the i-th fault phenomenon;

[0063] In an embodiment of the present invention, after receiving the fault record, the cloud server performs a time series analysis on the fault phenomenon according to the fault occurrence time, and obtains the time when the fault phenomenon first occurs and the time when the fault phenomenon last occurs.

[0064] Step S3: Calculate the fault phenomenon that will occur and the time to start predicting the fault phenomenon according to the first and last occurrence times of the acquired fault phenomenon;

[0065] The process of calculating the time threshold for the fault phenomenon to occur and starting to predict the fault phenomenon to occur in step S3 includes:

[0066] Step S3-1: The system obtains the current time T, and compares T with the set {(χ1, Tmin1, Tmax1), ..., (χ i , Tmin i , Tmax i ), …}, and extract the Tmin greater than T, and obtain the set {Tmin1, Tmin2, …, Tmin i}, compare the difference between Tmin and T in the set, and the Tmin with the smallest difference m Corresponding fault phenomenon χ m is the fault phenomenon to be predicted;

[0067] Step S3-2: Compare T with the set {(χ1, Tmin1, Tmax1), ..., (χ i , Tmin i , Tmax i ), …}, and extract the Tmax smaller than T to obtain the set {Tmax1, Tmax2, …, Tmax m}, compare the difference between Tmax and T in the set, and the Tmax with the smallest difference k is the fault phenomenon predicted last time k time of last appearance;

[0068] Step S3-3: Obtaining Tmax k With Tmin m At the middle moment, according to the formula:

[0069]

[0070] Wherein, ΔT represents the time threshold for starting fault prediction for the fault phenomenon that is about to occur;

[0071] In an embodiment of the present invention, the cloud server analyzes the first and last occurrence times of the acquired fault phenomenon, and calculates the time to start predicting the fault phenomenon that is about to occur.

[0072] Step S4: If the current time is the predicted time for the fault phenomenon to start, the system calculates the probability of each fault cause causing the fault phenomenon to occur;

[0073] The process in step S4 where the system calculates the probability of occurrence of each fault cause causing the fault phenomenon includes:

[0074] Step S4-1: Based on the historical fault phenomenon X = {χ1, χ2, ..., χ n} and historical fault causes M={m1,m2,…,m n}, extract all historical fault phenomena corresponding to the same fault cause, and form a set χ={χ1,χ2,…,χ k}, where χ is a subset of the set X;

[0075] Step S4-2: Calculate the same fault phenomenon x σ Under the condition of occurrence, a certain fault cause mi The probability of occurrence is based on the formula:

[0076]

[0077] in,

[0078] Among them, P(χ σ |m i ) is the fault cause m i Fault phenomenon under the conditions of occurrence σ Probability of occurrence; N χ The fault reason is m i The number of records of the corresponding fault phenomenon χ in X; N χσ The fault phenomenon is σ The number of records in the set χ; P(m i ) is among all the fault causes m i Probability of occurrence; N M is the number of records of all fault causes in set M; N mi The fault reason is m i The number of records in the set M; P(χ σ ) is the value of χ in all fault phenomena σ Probability of occurrence; N X is the number of records of all fault phenomena in X; N Xσ The fault phenomenon is σ The number of records in set X;

[0079] In an embodiment of the present invention, when the system detects that the current time is the time for fault prediction, the cloud server calculates the probability of the fault cause occurring under the condition that the fault phenomenon occurs.

[0080] Step S5: the system issues an early warning prompt according to the probability of occurrence of each fault cause leading to the fault phenomenon;

[0081] The process of the system performing an early warning prompt in step S5 includes:

[0082] Step S5-1: The system obtains the probability of each fault cause causing the fault phenomenon, and selects the fault cause with the highest probability through a comparator;

[0083] Step S5-2: The system detects the parameters of the communication device according to the cause of the fault. When it is identified that the parameters exceed the threshold range, a first-level warning is triggered, and the parameters are automatically adjusted. If they return to the threshold range, the first-level warning disappears;

[0084] Step S5-3: If the parameter cannot be restored to within the threshold range, the system issues a secondary warning and sends the cause of the fault to the application end connected to the communication device;

[0085] In an embodiment of the present invention, the cloud server uses a comparator to screen out the fault cause with the highest probability of occurrence. The fault cause causes a certain fault phenomenon in the communication equipment. The cloud server uploads the fault cause to the system. The system detects the fault cause and adjusts the abnormal parameters on the communication equipment end. If the parameter recovers to within the threshold range, the first-level warning disappears. If the parameter cannot recover to within the threshold range, the second-level warning will be triggered. The cloud server sends the fault cause to the application end to remind the staff to solve the fault cause.

[0086] In order to better implement the above method, an intelligent early warning system for power safety communication based on big data is also proposed. The early warning system includes an acquisition module, a transmission module, an analysis module, a fault prediction module, and an early warning module;

[0087] A collection module, used to collect each historical fault record in the communication device, wherein each fault record includes the fault phenomenon, occurrence time, and fault cause;

[0088] Transmission module, used for transmission of various information between modules;

[0089] An analysis module, which analyzes the occurrence time of the same fault phenomenon in all historical fault records to obtain the first and last occurrence time of the same fault phenomenon, and calculates the time to start predicting the fault phenomenon based on the first and last occurrence time of the fault phenomenon;

[0090] Among them, the analysis module includes a fault occurrence time analysis unit and a fault prediction time analysis unit;

[0091] A fault occurrence time analysis unit, used to sort the same fault phenomenon in all historical fault records according to the occurrence time, and obtain the first occurrence time and the last occurrence time of the same fault phenomenon;

[0092] A fault prediction time analysis unit, used to calculate the time to start predicting the fault phenomenon according to the acquired time of the first occurrence and the last occurrence of the fault phenomenon;

[0093] The fault prediction module is used to calculate the fault phenomenon that will occur and calculate the probability of each fault cause that causes the fault phenomenon to occur;

[0094] Among them, the fault prediction module includes a fault phenomenon prediction unit and a fault cause prediction unit;

[0095] A fault phenomenon prediction unit is used to analyze the time when the fault phenomenon first occurs and obtain the fault phenomenon to be predicted;

[0096] The fault cause prediction unit calculates the probability of a certain fault cause occurring when the same fault phenomenon occurs, based on the acquired number of historical fault phenomenon records and the number of historical fault cause records;

[0097] The early warning module is used to screen out the causes of the faults, detect the parameters of the communication equipment according to the causes of the faults, and give early warning prompts;

[0098] Among them, the early warning module includes a first-level early warning unit and a second-level early warning unit;

[0099] The first-level warning unit is used to screen out the cause of the fault and detect the parameters of the communication equipment according to the cause of the fault. If the parameter exceeds the threshold range, the first-level warning is triggered;

[0100] The secondary warning unit is used to identify parameters that cannot be restored to the threshold range and trigger the secondary warning.

[0101] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A power safety communication intelligent early warning method based on big data, characterized by: The early warning method comprises: Step S1: Collect each historical fault record in the communication device, wherein each historical fault record includes the fault phenomenon, occurrence time, and fault cause; Step S2: Analyze the occurrence time of the same fault phenomenon in all historical fault records to obtain the first occurrence time and the last occurrence time of the same fault phenomenon; Step S3: Calculate the fault phenomenon that will occur and predict the time when the fault phenomenon starts according to the first occurrence time and the last occurrence time of the fault phenomenon acquired; Step S4: If the current time is the time when the fault phenomenon starts to be predicted, the system calculates the probability of each fault cause causing the fault phenomenon to occur; Step S5: The system issues an early warning prompt based on the probability of occurrence of each fault cause that leads to the fault phenomenon.

2. According to the big data-based power safety communication intelligent early warning method of claim 1, it is characterized by: The process of obtaining the time when the fault phenomenon first occurs and the time when the fault phenomenon last occurs in step S2 includes: Step S2-1: Perform time series analysis on the occurrence time of the same fault phenomenon in all historical fault records to obtain the first occurrence time Tmin and the last occurrence time Tmax of the fault phenomenon; Step S2-2: The system integrates the acquired Tmin and Tmax to obtain a set {(χ1, Tmin1, Tmax1), ..., (χ i , Tmin i , Tmax i ), …}; Among them, χ i Indicates the i-th fault phenomenon.

3. According to the big data-based power safety communication intelligent early warning method of claim 1, it is characterized by: The process of calculating the time threshold of the fault phenomenon to be occurred and starting to predict the fault phenomenon to be occurred in step S3 includes: Step S3-1: The system obtains the current time T, and compares T with the set {(χ1, Tmin1, Tmax1), ..., (χ i , Tmin i , Tmax i ), …}, and extract the Tmin greater than T to obtain the set {Tmin1, Tmin2, …, Tmin i }, compare the difference between Tmin and T in the set, and the Tmin with the smallest difference m Corresponding fault phenomenon χ m is the fault phenomenon to be predicted; Step S3-2: Compare T with the set {(χ1, Tmin1, Tmax1), ..., (χ i , Tmin i , Tmax i ), …}, and extract the Tmax smaller than T to obtain the set {Tmax1, Tmax2, …, Tmax m }, compare the difference between Tmax and T in the set, and the Tmax with the smallest difference k is the fault phenomenon predicted last time k time of last appearance; Step S3-3: Obtaining Tmax k With Tmin m At the middle moment, according to the formula: Wherein, ΔT represents the time threshold for starting fault prediction for the fault phenomenon that is about to occur.

4. According to the big data-based power safety communication intelligent early warning method of claim 1, it is characterized by: The process in step S4 where the system calculates the probability of occurrence of each fault cause causing the fault phenomenon includes: Step S4-1: Based on the historical fault phenomenon X = {χ1, χ2, ..., χ n } and historical fault causes M={m1,m2,…,m n }, extract all historical fault phenomena corresponding to the same fault cause, and form a set χ={χ1,χ2,…,χ k }, where χ is a subset of the set X; Step S4-2: Calculate the same fault phenomenon x σ Under the condition of occurrence, a certain fault cause m i The probability of occurrence is based on the formula: in, Among them, P(χ σ |m i ) is the fault cause m i Fault phenomenon under the conditions of occurrence σ Probability of occurrence; N χ The fault reason is m i The number of records of the corresponding fault phenomenon χ in X; N χσ The fault phenomenon is σ The number of records in the set χ; P(m i ) is among all the fault causes m i Probability of occurrence; N M is the number of records of all fault causes in set M; N mi The fault reason is m i The number of records in the set M; P(χ σ ) is the value of χ in all fault phenomena σ Probability of occurrence; N X is the number of records of all fault phenomena in X; N Xσ The fault phenomenon is σ The number of records in set X.

5. According to the big data-based power safety communication intelligent early warning method of claim 1, it is characterized by: The process of the system performing early warning prompts in step S5 includes: Step S5-1: The system obtains the probability of each fault cause causing the fault phenomenon, and selects the fault cause with the highest probability through a comparator; Step S5-2: The system detects the parameters of the communication device according to the cause of the fault. When it is identified that the parameters exceed the threshold range, a first-level warning is triggered, and the parameters are automatically adjusted. If they return to the threshold range, the first-level warning disappears; Step S5-3: If the parameter cannot be restored to within the threshold range, the system issues a secondary warning and sends the cause of the fault to the application end connected to the communication device.

6. An intelligent early warning system for power safety communication based on big data, used to execute an intelligent early warning method for power safety communication based on big data as claimed in any one of claims 1 to 5, characterized in that: The early warning system includes: a collection module, a transmission module, an analysis module, a fault prediction module, and an early warning module; The acquisition module is used to collect each historical fault record in the communication device, and each fault record includes the fault phenomenon, occurrence time, and fault cause; The transmission module is used for transmitting various information between various modules; The analysis module analyzes the occurrence time of the same fault phenomenon in all historical fault records to obtain the first and last occurrence time of the same fault phenomenon, and calculates the time to start predicting the fault phenomenon based on the first and last occurrence time of the fault phenomenon. The fault prediction module is used to calculate the fault phenomenon that will occur and calculate the probability of each fault cause that causes the fault phenomenon to occur; The early warning module is used to screen out the fault cause that causes the fault phenomenon, detect the parameters of the communication device end according to the fault cause, and give an early warning prompt.

7. The power safety communication intelligent early warning system based on big data according to claim 6 is characterized by: The analysis module includes a fault occurrence time analysis unit and a fault prediction time analysis unit; The fault occurrence time analysis unit is used to sort the same fault phenomenon in all historical fault records according to the occurrence time, and obtain the first occurrence time and the last occurrence time of the same fault phenomenon; The fault prediction time analysis unit is used to calculate the time to start predicting the fault phenomenon according to the acquired time when the fault phenomenon first occurs and the time when the fault phenomenon last occurs.

8. The power safety communication intelligent early warning system based on big data according to claim 1 is characterized by: The fault prediction module includes a fault phenomenon prediction unit and a fault cause prediction unit; The fault phenomenon prediction unit is used to analyze the time when the fault phenomenon first occurs and obtain the fault phenomenon to be predicted; The fault cause prediction unit calculates the probability of a certain fault cause occurring when the same fault phenomenon occurs, based on the acquired number of historical fault phenomenon records and the number of historical fault cause records.

9. The big data-based power safety communication intelligent early warning system according to claim 6 is characterized by: The early warning module includes a primary early warning unit and a secondary early warning unit; The first-level warning unit is used to screen out the fault cause that causes the fault phenomenon, and detect the parameters of the communication device end according to the fault cause. If the parameter exceeds the threshold range, the first-level warning is triggered; The secondary warning unit is used to identify parameters that cannot be restored to a threshold range and trigger a secondary warning.

Images