Automatic fire extinguishing control method and device based on power data monitoring

Abstract

The invention relates to an automatic fire extinguishing control method and device based on power data monitoring, and relates to the technical field of power systems, and the method comprises the steps: monitoring an electrical parameter sequence, a vibration signal sequence and a temperature sequence of a power cable through a sensor in a power facility; performing contact resistance time domain fluctuation analysis and power consumption time domain fluctuation analysis to obtain a contact resistance fluctuation factor and a power consumption fluctuation factor, and calculating to obtain a temperature shock factor; according to the temperature sequence, change fitting verification of the temperature shock factor is carried out, and a verification risk factor is obtained; fire probability prediction is performed according to the electrical parameter sequence and the temperature sequence, fire factors are obtained through calculation in combination with verification risk factors, and when the fire factors are larger than a fire factor threshold value, a control signal is sent to control starting of a fire extinguishing device. The technical problems that in the prior art, an electric power facility cannot be effectively monitored and known in the early stage of fire breakout, the effectiveness of electric power data monitoring analysis is insufficient, and consequently fire hazards exist in the electric power facility are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular, to an automatic fire extinguishing control method and device based on power data monitoring. Background Art

[0002] During the operation of power facilities, problems such as varying degrees of aging, short circuits, and overloads exist in equipment, lines, etc., which are prone to causing fires. Among them, the temperature of a line may rise due to overload or poor temperature, which may then lead to the line catching fire. Without effective control, the fire spreads rapidly until it expands.

[0003] In the prior art, fire protection is generally achieved by monitoring voltage and current and using a circuit breaker protection device such as a fuse. However, some fuses cannot blow in time. Therefore, there are technical problems in the prior art that it is impossible to effectively monitor and obtain information in the early stage of a power facility catching fire, and the effectiveness of power data monitoring and analysis is insufficient, resulting in fire hazards in power facilities. Summary of the Invention

[0004] The present invention aims at the technical problems in the prior art that it is impossible to effectively monitor and obtain information in the early stage of a power facility catching fire, and the effectiveness of power data monitoring and analysis is insufficient, resulting in fire hazards in power facilities. Aiming at these problems, the present invention provides an automatic fire extinguishing control method and device based on power data monitoring.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] In a first aspect, the present invention provides an automatic fire extinguishing control method based on power data monitoring, including: monitoring the electrical parameter sequence, vibration signal sequence, and temperature sequence of a power cable through electrical sensors, vibration sensors, and temperature sensors in a power facility, and transmitting them to a cloud server through the Internet of Things; performing time-domain fluctuation analysis of the contact resistance based on the vibration signal sequence to obtain a contact resistance fluctuation factor, performing time-domain fluctuation analysis of power consumption based on the electrical parameter sequence to obtain a power consumption fluctuation factor, and calculating a temperature shock factor by combining the contact resistance fluctuation factor; performing change fitting verification of the temperature shock factor based on the temperature sequence to obtain a verification risk factor; predicting the fire probability based on the electrical parameter sequence and temperature sequence, and calculating a fire factor by combining the verification risk factor. When it is greater than a fire factor threshold, a control signal is sent to control the activation of a fire extinguishing device.

[0007] Second aspect, the present invention provides an automatic fire extinguishing control device based on power data monitoring, comprising: an electrical data monitoring module, configured to monitor the electrical parameter sequence, vibration signal sequence, and temperature sequence of a power cable through electrical sensors, vibration sensors, and temperature sensors within a power facility, and transmit them to a cloud server through the Internet of Things; a power impact analysis module, configured to perform time-domain fluctuation analysis of contact resistance based on the vibration signal sequence to obtain a contact resistance fluctuation factor, perform time-domain fluctuation analysis of power consumption based on the electrical parameter sequence to obtain a power consumption fluctuation factor, and calculate a temperature impact factor by combining the contact resistance fluctuation factor; a temperature change verification module, configured to perform change fitting verification of the temperature impact factor based on the temperature sequence to obtain a verification risk factor; a fire ignition prediction control module, configured to predict the fire ignition probability based on the electrical parameter sequence and temperature sequence, calculate a fire ignition factor by combining the verification risk factor, and send a control signal to control the activation of a fire extinguishing device when it is greater than a fire ignition factor threshold.

[0008] The beneficial effects of the present invention are as follows: By arranging sensors within the power facility, the present invention can monitor the electrical parameter sequence, vibration signal sequence, and temperature sequence of the power cable in real time, and transmit them to the cloud server through the Internet of Things, enabling remote data analysis and automated control, and avoiding the problem of out-of-control fires caused by the response lag or misoperation of traditional fuse protection. Among them, the monitoring and analysis of the vibration signal sequence can identify the looseness of the cable or joint, and calculate the contact resistance fluctuation factor through time-domain fluctuation analysis of the contact resistance, effectively identifying the resistance change caused by poor contact. The time-domain fluctuation analysis of power consumption extracts the power consumption fluctuation factor through the electrical parameter sequence, and calculates the temperature impact factor by combining the contact resistance fluctuation factor, accurately reflecting the instantaneous temperature impact phenomenon inside the power system caused by load fluctuations or poor contact, and discovering the risk of overload or short-term abnormal high temperature in advance. Further, the temperature impact factor is verified by fitting through the temperature sequence, verifying the occurrence times of the temperature impact, and calculating the verification risk factor. Finally, based on the verification risk factor, electrical parameter sequence, and temperature sequence, the fire ignition probability is predicted and the fire ignition factor is calculated, improving the prediction accuracy. When the fire ignition factor exceeds the set threshold, a control signal is automatically sent to trigger the fire extinguishing device, realizing the full-process intelligent control from fire risk identification to fire extinguishing response, being able to more accurately predict and respond to the fire hazards of power facilities, improving the monitoring effectiveness and fire extinguishing timeliness, and avoiding the development of the fire situation. Description of the Drawings

[0009] Figure 1 It is a schematic flowchart of an automatic fire extinguishing control method based on power data monitoring provided by the present invention;

[0010] Figure 2 It is a schematic structural diagram of an automatic fire extinguishing control device based on power data monitoring provided by the present invention.

[0011] Reference numerals: Electrical data monitoring module 11, power impact analysis module 12, temperature change verification module 13, fire prediction control module 14. Detailed implementation manners

[0012] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts fall within the protection scope of the present invention.

[0013] In the description of the present invention, the terms "first" and "second" are only used for descriptive purposes, and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the present invention, "a plurality" means two or more, unless otherwise specifically defined.

[0014] In the description of the present invention, the term "for example" is used to mean "serving as an example, illustration, or explanation". Any embodiment described as "for example" in the present invention is not necessarily construed as being more preferred or more advantageous than other embodiments. In order for any person skilled in the art to implement and use the present invention, the following description is given. In the following description, details are set forth for the purpose of explanation. It should be understood that those skilled in the art can recognize that the present invention can be implemented without using these specific details. In other instances, well-known structures and processes will not be elaborated in detail to avoid unnecessary details from obscuring the description of the present invention. Therefore, the present invention is not intended to be limited to the illustrated embodiments, but is consistent with the broadest scope that conforms to the principles and features disclosed in the present invention.

[0015] In the technical solution of the present invention, the collection and use of data are carried out with the permission of the user and comply with relevant regulations.

[0016] Embodiment 1, as Figure 1 shown, the embodiment of the present invention provides an automatic fire extinguishing control method based on power data monitoring. The method specifically includes the following steps:

[0017] S10: Monitor the electrical parameter sequence, vibration signal sequence, and temperature sequence of the power cable through electrical sensors, vibration sensors, and temperature sensors in the power facility, and transmit them to the cloud server through the Internet of Things.

[0018] In the embodiments of the present application, first, electrical sensors, vibration sensors, and temperature sensors arranged in power facilities are used to detect electrical parameters, vibration signals, and temperatures during the operation of the power facilities for subsequent fire analysis.

[0019] Among them, the acquired electrical parameter sequences, vibration signal sequences, and temperature sequences are transmitted to a cloud server through the Internet of Things for subsequent processing. More accurate processing and analysis can be provided through cloud computing power, and remote fire extinguishing analysis and control can be achieved.

[0020] Step S10 in the method provided by the embodiments of the present application includes:

[0021] Monitor the electrical parameters, vibration signals, and temperatures of the cables in the power facilities within the most recent preset time range at a preset frequency through electrical sensors, vibration sensors, and temperature sensors in the power facilities;

[0022] Arrange the electrical parameters, vibration signals, and temperatures according to the monitoring timestamp sequence within the preset time range to obtain electrical parameter sequences, vibration signal sequences, and temperature sequences, and transmit them to the cloud server through the Internet of Things.

[0023] In the embodiments of the present application, first, electrical sensors, vibration sensors, and temperature sensors are arranged in the power facilities to ensure comprehensive perception and monitoring of the operating states of the cables and their connecting components.

[0024] Exemplarily, the power facility is a power distribution cabinet, etc. The electrical sensor monitors electrical parameters such as the current and voltage of the cable line in the power facility. The vibration sensor monitors the vibration condition of the cable. For example, it is an acceleration sensor arranged on the cable. When the cable vibrates under an external force (such as the vibration when a heavy vehicle passes by), the vibration frequency and amplitude are monitored as vibration signals. The temperature sensor is, for example, a thermocouple that monitors the temperature on the surface of the cable line.

[0025] The electrical sensors, vibration sensors, and temperature sensors collect data on the operating states of the target cable within the most recent preset time range at a preset frequency (i.e., the set sampling interval, such as 1 s, 10 s, or 1 min) to obtain electrical parameters, vibration signals, and temperatures. The most recent preset time range is, for example, the most recent 10 min. The time window mechanism is adopted to record data within the most recent preset time range to ensure the timeliness and integrity of the data, and at the same time, it can also avoid waste of storage and computing resources. For example, 10 electrical parameters within the most recent 10 s are collected.

[0026] Furthermore, to ensure data consistency and traceability, each piece of collected data is attached with a timestamp to mark the specific collection time point of the data, forming a monitoring timestamp sequence within a recent preset time range. According to the monitoring timestamp sequence, the monitored electrical parameters, vibration signals, and temperature are arranged in time series to form a complete electrical parameter sequence, vibration signal sequence, and temperature sequence.

[0027] The electrical parameter sequence can be used to analyze the fluctuations of current and voltage. When the cable is vibrated, it will cause the joints to loosen, resulting in an increase in contact resistance. The vibration signal sequence can be used to analyze the fluctuations of the contact resistance, and the temperature sequence can be used to track the temperature rise trend of the cable and equipment.

[0028] After completing data collection and time series arrangement, through the Internet of Things, the electrical parameter sequence, vibration signal sequence, and temperature sequence are uploaded to the cloud server. For example, a gateway device can be set up in the power facility for data transmission, sending the monitored data to the cloud server for processing or receiving signals sent by the cloud server, avoiding consuming computing power or affecting efficiency in local processing. The cloud server receives the above data and performs subsequent processing.

[0029] The embodiment of the present application provides basic data support for the operation state assessment of power facilities, fire risk prediction, and automatic fire extinguishing control by monitoring multi-dimensional data. This multi-modal monitoring method is more real-time, accurate, and predictive compared to traditional single current and voltage monitoring schemes, and can effectively improve the safety and intelligent level of the power system.

[0030] S20: According to the vibration signal sequence, perform time-domain fluctuation analysis of the contact resistance to obtain a contact resistance fluctuation factor. According to the electrical parameter sequence, perform time-domain fluctuation analysis of power consumption to obtain a power consumption fluctuation factor. Combine the contact resistance fluctuation factor to calculate and obtain a temperature shock factor.

[0031] In the embodiment of the present application, the vibration of the cable line in the power facility will cause the joints to loosen, which will in turn cause an increase in contact resistance and fluctuations. When the contact resistance increases, the thermal effect of the joints increases and the temperature rises, which may lead to a fire. Therefore, according to the vibration signal sequence, perform time-domain fluctuation analysis of the contact resistance to obtain a contact resistance fluctuation factor for subsequent fire prediction analysis.

[0032] Moreover, the fluctuations of electrical parameters in the power facility will also cause an increase in thermal effect and temperature rise, which may lead to a fire. For example, instantaneous large fluctuations in current and voltage. Therefore, according to the electrical parameter sequence, perform time-domain fluctuation analysis of power consumption to obtain a power consumption fluctuation factor.

[0033] Finally, combine the two dimensions to calculate the temperature shock factor to reflect the degree of temperature change caused by the vibration of the cable and the fluctuations of electrical parameters.

[0034] Step S20 in the method provided by the embodiments of this application includes:

[0035] Within the vibration signal sequence, an amplitude signal sequence is extracted and obtained;

[0036] According to the vibration monitoring data of the power equipment, a set of sample amplitude signal sequences is collected, and the change amplitude of the contact resistance before and after vibration according to different sample amplitude signal sequences is collected and labeled as a set of sample contact resistance fluctuation factors;

[0037] Using the set of sample amplitude signal sequences and the set of sample contact resistance fluctuation factors as supervised training data and test data, machine learning is used to train and test the contact resistance time-domain fluctuation analysis path, and stop after convergence;

[0038] Input the amplitude signal sequence into the contact resistance time-domain fluctuation analysis path, and output and obtain the contact resistance fluctuation factor.

[0039] In the embodiments of this application, in order to accurately identify the contact resistance fluctuation of the cable joints in the power facilities, first, the vibration signal sequence is processed to extract the amplitude signal sequence therein. Specifically, the mechanical vibration amplitude in the vibration signal is extracted, and can be obtained through acceleration vibration sensors for extraction and processing. For example, the amplitude of a certain cable joint at a certain moment is 0.08g (g is the unit of gravitational acceleration). In this way, the amplitude signal sequence within the vibration signal sequence is extracted and obtained.

[0040] Furthermore, after the amplitude signal extraction is completed, according to the vibration monitoring data of the power equipment, a set of sample amplitude signal sequences is collected, that is, the sample amplitude signal sequences of the cables monitored and extracted in the same type of power facilities at different times. Then, the change amplitude of the contact resistance before and after vibration according to different sample amplitude signal sequences is collected. The contact resistance before and after vibration can be measured by a micro-ohmmeter. Calculate the ratio of the contact resistance after vibration to the contact resistance before vibration as the contact resistance change amplitude, and then label it as the sample contact resistance fluctuation factor. For example, the contact resistance of a certain cable joint before and after vibration is 0.02 mΩ, and the contact resistance after vibration is 0.5 mΩ, then the resistance change amplitude is 0.5 mΩ / 0.02 mΩ = 250%. In this way, a set of sample contact resistance fluctuation factors is collected and labeled.

[0041] Furthermore, using the set of sample amplitude signal sequences and the set of sample contact resistance fluctuation factors as supervised training data and test data, machine learning is used to train the contact resistance time-domain fluctuation analysis path for analyzing the contact resistance time-domain fluctuation and conduct tests until it converges.

[0042] Exemplarily, a BP neural network is used to construct an analysis path for the time-domain fluctuation of contact resistance. Its specific architecture includes an input layer, an output layer, and hidden layers. The hidden layers include two layers, and the activation function is the ReLU activation function. The sample amplitude signal sequence set and the sample contact resistance fluctuation factor set are divided into supervised training data and test data according to a ratio of 8:2. During the training process, the sample amplitude signal sequence is input, and the output contact resistance fluctuation factor is obtained. The mean square error (MSE) loss function is used to calculate the error loss with the corresponding sample contact resistance fluctuation factor, and then error backpropagation is performed to update the network weights. Iterative training is carried out in this way until the error loss is less than a set threshold, for example, less than 10 −4 , and then testing is performed. If the test error loss is also less than the set threshold, the training converges; otherwise, iterative training continues until convergence.

[0043] The current amplitude signal sequence is input into the trained analysis path for the time-domain fluctuation of contact resistance, and the corresponding contact resistance fluctuation factor is obtained as the output. The contact resistance fluctuation factor reflects the current amplitude of the change in contact resistance caused by the loosening of the cable line joints due to vibration.

[0044] In the embodiment of the present application, by monitoring vibration signals and analyzing the fluctuation of contact resistance, the change of contact resistance can be obtained, so as to realize the intelligent identification and early warning of the poor contact state of power equipment, and provide accurate data support for fire protection and equipment operation and maintenance.

[0045] The step S20 in the method provided in the embodiment of the present application further includes:

[0046] Randomly select multiple groups of current parameters within the electrical parameter sequence, and each group of current parameters includes two current parameters;

[0047] Calculate the current fluctuation amplitude between the two current parameters in each group of current parameters as multiple electrical fluctuation coefficients;

[0048] Screen the largest electrical fluctuation coefficient as the power consumption fluctuation factor.

[0049] In the embodiment of the present application, when the user's power consumption suddenly fluctuates, it will cause fluctuations in electrical parameters, and the fluctuations in electrical parameters will also cause the temperature to rise, which may lead to a fire in the line. Therefore, an analysis of the electrical fluctuations in the recent power facilities is carried out based on the electrical parameter sequence.

[0050] Specifically, an electrical parameter sequence refers to a set of data such as current, voltage, power factor, etc. that changes over time. In the embodiments of this application, electrical fluctuation analysis is performed based on the current. Specifically, a current parameter sequence is extracted from the electrical parameter sequence, and then multiple groups of current parameter pairs are randomly selected from the electrical parameter sequence. Each group of data contains current values at two different time points, such as (I1, I2), (I3, I4),..., (Im, In). For example, 5 groups of current parameter pairs are randomly selected, and m and n are 9 and 10.

[0051] Furthermore, calculate the current fluctuation amplitude between the two current parameters within each group of current parameters as multiple electrical fluctuation coefficients. Specifically, calculate the ratio of the larger current parameter to the smaller current parameter as the electrical fluctuation coefficient. For example, if Im and In are 100 A and 75 A respectively, then the electrical fluctuation coefficient is 100 A / 75 A = 133%. In this way, multiple electrical fluctuation coefficients are calculated.

[0052] Furthermore, screen and select the largest electrical fluctuation coefficient among the multiple electrical fluctuation coefficients as the power consumption fluctuation factor, which reflects the maximum fluctuation degree of the current within a preset time range recently. The larger the power consumption fluctuation factor, the greater the probability of a fire occurrence.

[0053] In step S20 of the method provided in the embodiments of this application, combine the power consumption fluctuation factor and the contact resistance fluctuation factor to calculate the temperature shock factor as follows:

[0054] ;

[0055] Wherein, is the temperature shock factor, C is the heat dissipation factor, which depends on the heat dissipation capacity of the power facilities, ambient temperature, ventilation conditions, etc. For example, when the ventilation environment is good and the ambient temperature is less than 25 °C, C is set to 0.6. When the ventilation is poor and the ambient temperature is greater than 25 °C and less than 35 °C, C is set to 1. When the ambient temperature is greater than 35 °C, C is set to 1.5. is the power consumption fluctuation factor, is the contact resistance fluctuation factor.

[0056] In the embodiments of this application, in order to accurately evaluate the instantaneous shock impact on the temperature of the power facilities during recent operation, combine the contact resistance fluctuation factor of the contact resistance fluctuation and the power consumption fluctuation factor of the power consumption current fluctuation to analyze and calculate the temperature shock factor.

[0057] Among them, the law of temperature change in the power facilities is based on the Joule heat effect, and there is the following formula:

[0058] ; where Q is the heat, I is the current, R is the contact resistance, and t is the time. Therefore, the greater the power consumption fluctuation factor, the greater the contact resistance fluctuation factor, that is, the greater the amplitude of the current fluctuation caused by the power consumption fluctuation, the greater the fluctuation of the contact resistance due to vibration, and the greater the temperature fluctuation of the cable in the power facility. The temperature shock factor reflects the amplitude of the change in the cable temperature affected by the power consumption fluctuation factor and the contact resistance fluctuation factor, and is proportional to the square of the power consumption fluctuation factor and the contact resistance fluctuation factor.

[0059] In the embodiment of the present application, by analyzing the current fluctuation and the amplitude of the contact resistance fluctuation based on the recent electrical parameter sequence and vibration signal sequence, and then calculating the amplitude of the future temperature fluctuation, that is, the temperature shock factor. Since there is a certain hysteresis in the cable heating, the amplitude of the possible future temperature fluctuation caused by the recent electrical parameter sequence and vibration signal sequence can be predicted and calculated through the above steps, thereby providing a data basis for the subsequent analysis and calculation of the probability of power facility fire. Specifically, the greater the temperature shock factor, the greater the probability of fire.

[0060] S30: According to the temperature sequence, perform the change fitting verification of the temperature shock factor to obtain the verification risk factor.

[0061] In the embodiment of the present application, according to the temperature sequence and the temperature monitoring data of the cable in the power facility within the historical time, the change fitting verification of the currently predicted and calculated temperature shock factor is performed. Among them, specifically analyze the proportion of the temperature shock factor in the cable temperature monitoring within the historical time.

[0062] If the proportion of the temperature shock factor is high, it means that the amplitude of the currently predicted and calculated cable temperature change occurs more frequently within the historical time, is relatively safe, and the probability of fire is low. If the proportion of the temperature shock factor is low, it means that the amplitude of the currently predicted and calculated cable temperature change occurs less frequently within the historical time, is relatively unsafe, belongs to an extreme situation, and the probability of fire is high.

[0063] The step S30 in the method provided by the embodiment of the present application includes:

[0064] Retrieve the historical temperature matrix within the historical time and add the temperature sequence;

[0065] In the historical temperature matrix, extract the number of time windows within the preset time window where the temperature change amplitude is greater than or equal to the temperature shock factor, calculate the time length ratio, and calculate to obtain the verification risk factor.

[0066] In the embodiment of the present application, in the data record of the cloud server, retrieve all the temperature sequences transmitted by the power setting cable monitoring within the historical time, sort them according to time, and obtain the historical temperature matrix.

[0067] Then configure a preset time window, which is the duration length of the cable temperature fluctuation in the preset time window. Since the cable temperature change has a certain hysteresis compared to the current change, a preset time window is set to verify the temperature shock factor. Specifically, calculate the maximum time length from the start of the cable temperature fluctuation to its decline within the historical time in the power facility as the preset time window. For example, it is 10 min.

[0068] Furthermore, in the historical temperature matrix, divide the historical temperature matrix according to the preset time window to obtain historical continuous temperature sequences within multiple preset time windows. Then extract the number of time windows in which the temperature change amplitude within the preset time window is greater than or equal to the temperature shock factor. For example, calculate the ratio of the maximum temperature to the minimum temperature in the historical continuous temperature sequence within each preset time window, and then determine whether it is greater than or equal to the temperature shock factor. If so, record it as a time window with a temperature change amplitude greater than or equal to the temperature shock factor. In this way, the number of time windows is obtained through discrimination and statistics.

[0069] Furthermore, calculate the ratio of the number of these time windows to the total number of preset time windows divided in the historical temperature matrix as the time length ratio. Further, subtract the time length ratio from 1 to obtain the verification risk factor. That is, the fewer the number of times the amplitude of the cable temperature change (temperature shock factor) calculated by the current prediction appears in the historical time, the more extreme the current temperature change amplitude is, and the greater the risk, so the verification risk factor is larger. Exemplarily, if the time length ratio is 3%, the verification risk factor is 1 - 3% = 97%.

[0070] Among them, if the occurrence ratio of the temperature shock factor in the historical temperature matrix is high, that is, the verification risk factor is small, it indicates that the amplitude of the cable temperature change calculated by the current prediction occurs more frequently in the historical time, is relatively safe, and the probability of fire occurrence is low. If the occurrence ratio of the temperature shock factor in the historical temperature matrix is low, that is, the verification risk factor is large, it indicates that the amplitude of the cable temperature change calculated by the current prediction occurs less frequently in the historical time, is relatively unsafe, belongs to an extreme situation, and the probability of fire occurrence is high.

[0071] By retrieving and verifying the temperature shock factor in the historical data of the power facility, obtaining the time ratio of its occurrence in the temperature fluctuation change in the historical time, and calculating the verification risk factor, the risk degree and the probability of fire occurrence of the temperature shock factor calculated by the current prediction can be reflected. By introducing the verification of power consumption fluctuation, contact resistance fluctuation and temperature shock fluctuation, the accuracy and reliability of the fire prediction analysis of the power facility are improved.

[0072] S40: Predict the fire probability based on the electrical parameter sequence and the temperature sequence, calculate the fire factor by combining the verification risk factor, and when it is greater than the fire factor threshold, send a control signal to control the activation of the fire extinguishing device.

[0073] In the embodiment of the present application, machine learning is used to predict the fire probability based on the electrical parameter sequence and the temperature sequence. Combining the verification risk factor calculated by the temperature fluctuation analysis in the foregoing content, a comprehensive fire factor is further calculated to reflect the probability of a fire occurring in the power facility. Then, it is determined whether to perform automatic fire extinguishing. Specifically, it is judged whether the fire factor is greater than the fire factor threshold. If so, a control signal is sent to control the activation of the fire extinguishing device. If not, no processing is performed and the monitoring continues.

[0074] Step S40 in the method provided by the embodiment of the present application includes:

[0075] Train a fire probability predictor;

[0076] Input the electrical parameter sequence and the temperature sequence into the fire probability predictor to obtain the predicted fire probability;

[0077] Calculate the fire factor based on the fire probability and the verification risk factor, and judge whether the fire factor is greater than the fire factor threshold. If so, send a control signal to control the activation of the fire extinguishing device. If not, no processing is performed.

[0078] In the embodiment of the present application, first, based on machine learning, a fire probability predictor for predicting the fire probability based on two-dimensional data of the electrical parameter sequence and the temperature sequence is trained. Among them, when there are electrical changes and a sharp rise in temperature, it will cause a fire in the cable, and the greater the electrical change and the greater the temperature rise, the greater the fire probability of the cable. Therefore, the fire probability can be predicted based on the electrical parameter sequence and the temperature sequence.

[0079] Training the fire probability predictor includes:

[0080] According to the fire monitoring data of similar power facilities, collect a set of sample electrical parameter sequences and a set of sample temperature sequences, and collect the proportion of fires under different sample electrical parameter sequences and sample temperature sequences, which is labeled as a set of sample fire probabilities;

[0081] Use the set of sample electrical parameter sequences, the set of sample temperature sequences, and the set of sample fire probabilities to train and test the fire probability predictor until convergence.

[0082] In the embodiments of the present application, according to the fire monitoring data of similar power facilities, the monitored sample electrical parameter sequences and sample temperature sequences are collected to obtain a sample electrical parameter sequence set and a sample temperature sequence set. Then, the proportions of fire occurrence events under different sample electrical parameter sequences and sample temperature sequences are collected and labeled as a sample fire probability set. Exemplarily, the fire probability is 1%.

[0083] Among them, the number of power facilities catching fire under multiple sample electrical parameter sequences and sample temperature sequences with the same average electrical parameter (such as average current) and average temperature can be calculated, and then the ratio of the number of multiple sample electrical parameter sequences and sample temperature sequences with the same average electrical parameter (such as average current) and average temperature is calculated as the fire probability.

[0084] In the embodiments of the present application, the sample electrical parameter sequence set, the sample temperature sequence set, and the sample fire probability set are continuously used for the training and testing of the fire probability predictor. For example, a BP neural network is used to construct the fire probability predictor, and its specific architecture includes an input layer, an output layer, and a hidden layer. The hidden layer includes three layers, and the activation function is the ReLU activation function. The input features of the input layer are the sample electrical parameter sequences and sample temperature sequences, and the output features of the output layer are the sample fire probabilities. The sample electrical parameter sequence set, the sample temperature sequence set, and the sample fire probability set are divided into supervised training data and test data according to a ratio of 8:2. During the training process, the sample electrical parameter sequences and sample temperature sequences are input, the obtained fire probability is obtained, the mean square error (MSE) loss function is used to calculate the error loss with the corresponding sample fire probability, and then error backpropagation is performed to update the network weights. Such iterative training is carried out until the error loss is less than a set threshold, such as less than 10 −4 , and then testing is performed. If the test error loss is also less than the set threshold, the training converges; otherwise, iterative training continues until convergence.

[0085] Based on the trained fire probability predictor, the currently collected electrical parameter sequence and temperature sequence are input into the fire probability predictor, and the predicted output obtains the fire probability.

[0086] Furthermore, based on the verification risk factor calculated in the foregoing content, combined with the fire probability, a parameter that comprehensively reflects the fire risk probability is calculated based on two dimensions, that is, the fire factor. For example, the mean value of the fire probability and the verification risk factor is calculated as the fire factor. If the fire probability is 1% and the verification risk factor is 3%, then the fire factor is (1% + 3%) / 2 = 2%.

[0087] Further, it is determined whether the fire - starting factor is greater than the fire - starting factor threshold. If so, the probability of the power facility cable catching fire is extremely high, and automatic fire extinguishing is required to avoid the development of the fire due to undetected fire extinguishing and extinguish the fire in the early stage. Conversely, if not, no treatment is carried out and the monitoring continues. The fire - starting factor threshold can be set by processing the electrical parameter sequence, vibration signal sequence, and temperature sequence obtained from the monitoring of multiple similar power facilities before catching fire using the methods described above to obtain the fire - starting factor and calculating the mean value. For example, the fire - starting factor threshold is 85%.

[0088] Among them, if the fire - starting factor is greater than the fire - starting factor threshold, the cloud server can send a control signal to start the fire - extinguishing device in the power facility through the Internet of Things for fire extinguishing. The fire - extinguishing device is, for example, an aerosol fire - extinguishing device. The control signal can be a thermal - sensitive wire, which controls the burning of the medicine column in the aerosol fire - extinguishing device through an electric circuit and then starts to spray aerosol for fire extinguishing.

[0089] By fusing and verifying the risk factor and predicting the fire - starting probability to calculate the fire - starting factor, and fusing the fire - starting risk probability parameters obtained from processing two dimensions, the accuracy and reliability of fire - starting prediction can be improved, thereby enhancing the stability of fire - extinguishing control and the safety of power facilities.

[0090] The automatic fire - extinguishing control method based on power data monitoring provided by the embodiments of the present invention has at least the following technical effects:

[0091] In the embodiments of the present invention, by arranging sensors in power facilities, real - time monitoring of the electrical parameter sequence, vibration signal sequence, and temperature sequence of power cables is carried out and transmitted to the cloud server through the Internet of Things, which can realize remote data analysis and automatic control, and avoid the problem of out - of - control fires caused by the lag or misoperation of traditional fuse protection. Among them, the monitoring and analysis of the vibration signal sequence can identify the looseness of cables or joints, calculate the contact - resistance fluctuation factor by using the time - domain fluctuation analysis of contact resistance, and effectively identify the resistance change caused by poor contact. The power - time - domain fluctuation analysis extracts the power - consumption fluctuation factor through the electrical parameter sequence, combines the contact - resistance fluctuation factor to calculate the temperature - shock factor, accurately reflects the instantaneous temperature - shock phenomenon inside the power system caused by load fluctuation or poor contact, and discovers the risk of overload or short - term abnormal high temperature in advance. Further, the temperature - shock factor is fitted and verified through the temperature sequence to verify the occurrence times of temperature shock and calculate the verification risk factor. Finally, based on the verification risk factor, electrical parameter sequence, and temperature sequence, the fire - starting probability is predicted and the fire - starting factor is calculated to improve the prediction accuracy. When the fire - starting factor exceeds the set threshold, a control signal is automatically sent to trigger the fire - extinguishing device, realizing the full - process intelligent control from fire - risk identification to fire - extinguishing response, being able to more accurately predict and respond to the fire hazards of power facilities, enhancing the monitoring effectiveness and fire - extinguishing timeliness, and avoiding the development of the fire situation.

[0092] Example 2, as Figure 2 shown, with the same inventive concept as the automatic fire extinguishing control method based on power data monitoring in Example 1, the embodiment of the present invention also provides an automatic fire extinguishing control device based on power data monitoring. The explanation of the automatic fire extinguishing control method based on power data monitoring in Example 1 also applies to an automatic fire extinguishing control device based on power data monitoring. The device includes:

[0093] The electrical data monitoring module 11 is used to monitor the electrical parameter sequence, vibration signal sequence and temperature sequence of the power cable through electrical sensors, vibration sensors and temperature sensors in the power facilities, and transmit them to the cloud server through the Internet of Things;

[0094] The power impact analysis module 12 is used to perform time-domain fluctuation analysis of the contact resistance according to the vibration signal sequence to obtain the contact resistance fluctuation factor, perform time-domain fluctuation analysis of the power consumption according to the electrical parameter sequence to obtain the power consumption fluctuation factor, and calculate the temperature impact factor by combining the contact resistance fluctuation factor;

[0095] The temperature change verification module 13 is used to perform change fitting verification of the temperature impact factor according to the temperature sequence to obtain the verification risk factor;

[0096] The fire ignition prediction control module 14 is used to predict the fire ignition probability according to the electrical parameter sequence and temperature sequence, calculate the fire ignition factor by combining the verification risk factor, and send a control signal to control the start of the fire extinguishing device when it is greater than the fire ignition factor threshold.

[0097] Furthermore, the electrical data monitoring module 11 is further used for:

[0098] Monitoring the electrical parameters, vibration signals and temperature of the cables in the power facilities through electrical sensors, vibration sensors and temperature sensors in the power facilities at a preset frequency within a preset recent time range;

[0099] Arranging the electrical parameters, vibration signals and temperature according to the monitoring timestamp sequence within the preset time range to obtain the electrical parameter sequence, vibration signal sequence and temperature sequence, and transmitting them to the cloud server through the Internet of Things.

[0100] Furthermore, the power impact analysis module 12 is further used for:

[0101] Extracting the amplitude signal sequence within the vibration signal sequence;

[0102] Collecting a set of sample amplitude signal sequences according to the vibration monitoring data of the power equipment, and collecting the change amplitude of the contact resistance before and after the vibration according to different sample amplitude signal sequences, and labeling it as a set of sample contact resistance fluctuation factors;

[0103] Using the set of sample amplitude signal sequences and the set of sample contact resistance fluctuation factors as supervised training data and test data, and using machine learning to train and test the contact resistance time-domain fluctuation analysis path, and stopping after convergence;

[0104] Input the amplitude signal sequence into the contact resistance time-domain fluctuation analysis path, and output to obtain the contact resistance fluctuation factor.

[0105] Furthermore, the power impact analysis module 12 is further configured to:

[0106] Randomly select multiple groups of current parameters within the electrical parameter sequence, and each group of current parameters includes two current parameters;

[0107] Calculate the current fluctuation amplitude between the two current parameters in each group of current parameters as multiple electrical fluctuation coefficients;

[0108] Screen the largest electrical fluctuation coefficient as the power consumption fluctuation factor.

[0109] Furthermore, the power impact analysis module 12 is further configured to:

[0110] Calculate the temperature impact factor in combination with the contact resistance fluctuation factor, as shown in the following formula:

[0111] ;

[0112] Wherein, is the temperature impact factor, C is the heat dissipation factor, is the power consumption fluctuation factor, is the contact resistance fluctuation factor.

[0113] Furthermore, the temperature change verification module 13 is further configured to:

[0114] Retrieve the historical temperature matrix within the historical time and add the temperature sequence;

[0115] Within the historical temperature matrix, extract the number of time windows with a temperature change amplitude greater than or equal to the temperature impact factor within a preset time window, calculate the time length ratio, and calculate the obtained verification risk factor.

[0116] Furthermore, the fire ignition prediction control module 14 is further configured to:

[0117] Train the fire ignition probability predictor;

[0118] Input the electrical parameter sequence and the temperature sequence into the fire ignition probability predictor to predict the obtained fire ignition probability;

[0119] According to the fire probability and the verification risk factor, calculate the fire factor, and determine whether the fire factor is greater than the fire factor threshold. If so, send a control signal to control the activation of the fire extinguishing device; if not, do not perform any processing.

[0120] Among them, training the fire probability predictor includes:

[0121] According to the fire monitoring data of similar power facilities, collect the sample electrical parameter sequence set and the sample temperature sequence set, and collect the proportion of fires under different sample electrical parameter sequences and sample temperature sequences, and label it as the sample fire probability set;

[0122] Use the sample electrical parameter sequence set, the sample temperature sequence set, and the sample fire probability set to train and test the fire probability predictor until convergence.

[0123] It should be noted that in the above embodiments, the descriptions of each embodiment have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0124] Those skilled in the art should understand that the embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0125] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0126] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means implements the functions in the flow Figure 1 one flow or multiple flows and / or blocksFigure 1 The functions specified in one or more boxes.

[0127] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one Figure 1 flow or multiple flows and / or boxes Figure 1 flow or multiple flows and / or the functions specified in one or more boxes.

[0128] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications to these embodiments once they learn the basic creative concept.

[0129] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. An automatic fire extinguishing control method based on power data monitoring, characterized in that, The method includes: Monitoring the electrical parameter sequence, vibration signal sequence, and temperature sequence of the power cable through electrical sensors, vibration sensors, and temperature sensors within the power facility, and transmitting them to the cloud server through the Internet of Things; Performing time-domain fluctuation analysis of the contact resistance based on the vibration signal sequence to obtain the contact resistance fluctuation factor, performing time-domain fluctuation analysis of power consumption based on the electrical parameter sequence to obtain the power consumption fluctuation factor, and calculating the temperature shock factor by combining the contact resistance fluctuation factor; Performing change fitting verification of the temperature shock factor based on the temperature sequence to obtain the verification risk factor; Predicting the fire probability based on the electrical parameter sequence and temperature sequence, calculating the fire factor by combining the verification risk factor, and sending a control signal to control the activation of the fire extinguishing device when it is greater than the fire factor threshold.

2. The automatic fire extinguishing control method based on power data monitoring according to claim 1, wherein Monitoring the electrical parameter sequence, vibration signal sequence, and temperature sequence of the power cable through electrical sensors, vibration sensors, and temperature sensors within the power facility, and transmitting them to the cloud server through the Internet of Things, including: Monitoring the electrical parameters, vibration signals, and temperature of the cable within the power facility at a preset frequency through electrical sensors, vibration sensors, and temperature sensors within the power facility for a recently preset time range; Arranging the electrical parameters, vibration signals, and temperature according to the monitoring timestamp sequence within the preset time range to obtain the electrical parameter sequence, vibration signal sequence, and temperature sequence, and transmitting them to the cloud server through the Internet of Things.

3. The automatic fire extinguishing control method based on power data monitoring according to claim 1, wherein, Performing time-domain fluctuation analysis of the contact resistance based on the vibration signal sequence to obtain the contact resistance fluctuation factor, including: Extracting the amplitude signal sequence within the vibration signal sequence; Collecting a set of sample amplitude signal sequences based on the vibration monitoring data of the power equipment, and collecting the change amplitude of the contact resistance before and after vibration according to different sample amplitude signal sequences, and labeling them as a set of sample contact resistance fluctuation factors; Using the set of sample amplitude signal sequences and the set of sample contact resistance fluctuation factors as supervised training data and test data, and using machine learning to train and test the time-domain fluctuation analysis path of the contact resistance, and stopping after convergence; Inputting the amplitude signal sequence into the time-domain fluctuation analysis path of the contact resistance, and outputting the contact resistance fluctuation factor.

4. The automatic fire extinguishing control method based on power data monitoring according to claim 1, wherein, Performing time-domain fluctuation analysis of power consumption based on the electrical parameter sequence to obtain the power consumption fluctuation factor, including: Randomly selecting multiple groups of current parameters within the electrical parameter sequence, and each group of current parameters includes two current parameters; Calculating the current fluctuation amplitude between the two current parameters in each group of current parameters as multiple electrical fluctuation coefficients; Selecting the largest electrical fluctuation coefficient as the power consumption fluctuation factor.

5. The automatic fire extinguishing control method based on power data monitoring according to claim 1, wherein Calculating the temperature shock factor by combining the contact resistance fluctuation factor, as shown in the following formula: ； Among them, is the temperature shock factor, C is the heat dissipation factor, is the power consumption fluctuation factor, is the contact resistance fluctuation factor.

6. The automatic fire extinguishing control method based on power data monitoring according to claim 1, characterized in that, Performing change fitting verification of the temperature shock factor based on the temperature sequence to obtain the verification risk factor, including: Retrieving the historical temperature matrix within the historical time and adding the temperature sequence; Extracting the number of time windows within the historical temperature matrix where the temperature change amplitude within a preset time window is greater than or equal to the temperature shock factor, calculating the time length ratio, and calculating the verification risk factor.

7. The automatic fire extinguishing control method based on power data monitoring according to claim 1, wherein Based on the verification risk factor, fire probability prediction is performed according to the electrical parameter sequence and temperature sequence to obtain the fire probability, including: Training a fire probability predictor; Inputting the electrical parameter sequence and temperature sequence into the fire probability predictor to predict and obtain the fire probability; Calculating a fire factor based on the fire probability and the verification risk factor, and determining whether the fire factor is greater than the fire factor threshold. If so, sending a control signal to control the activation of the fire extinguishing device; if not, no processing is performed.

8. The automatic fire extinguishing control method based on power data monitoring according to claim 7, characterized in that, Training the fire probability predictor includes: According to the fire monitoring data of similar power facilities, collecting a set of sample electrical parameter sequences and a set of sample temperature sequences, and collecting the proportion of fires under different sample electrical parameter sequences and sample temperature sequences, which is labeled as a set of sample fire probabilities; Using the set of sample electrical parameter sequences, the set of sample temperature sequences, and the set of sample fire probabilities to train and test the fire probability predictor until convergence.

9. An automatic fire extinguishing control device based on power data monitoring, characterized in that, The device is used to execute the method according to any one of claims 1-8, and the device includes: An electrical data monitoring module, configured to monitor the electrical parameter sequence, vibration signal sequence, and temperature sequence of the power cable through electrical sensors, vibration sensors, and temperature sensors in the power facility, and transmit them to the cloud server through the Internet of Things; A power impact analysis module, configured to perform time-domain fluctuation analysis of the contact resistance based on the vibration signal sequence to obtain a contact resistance fluctuation factor, perform time-domain fluctuation analysis of the power consumption based on the electrical parameter sequence to obtain a power consumption fluctuation factor, and calculate a temperature impact factor by combining the contact resistance fluctuation factor; A temperature change verification module, configured to perform change fitting verification of the temperature impact factor according to the temperature sequence to obtain a verification risk factor; A fire prediction control module, configured to predict the fire probability according to the electrical parameter sequence and temperature sequence, calculate a fire factor by combining the verification risk factor, and send a control signal to control the activation of the fire extinguishing device when it is greater than the fire factor threshold.

Images