A gas monitoring method and system for power distribution room
By using gas sensors and gas phase concentration prediction models in the distribution room, combined with environmental parameters to predict gas changes, the problems of poor detection error and predictability in high-temperature and humid environments in the existing technology are solved, and gas abnormalities in the distribution room are achieved High precision to make advance judgment and quick response.
Patent Information
- Application Number
- CN202411194872.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-29
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-08-29
AI Technical Summary
The existing gas monitoring technology in the distribution room has detection errors and poor foresight in high temperature and humid environments, and it is impossible to effectively judge gas abnormalities in advance, resulting in equipment failures and safety hazards.
The gas sensor combined with the gas phase concentration prediction model is used to predict the gas phase concentration change law of the target gas by monitoring environmental parameters such as temperature and humidity, generate a relationship curve, judge the gas abnormality in advance, and compare it when access control and ventilation conditions change to determine the gas Abnormal concentration.
It improves the accuracy and foresight of gas monitoring in the distribution room, and can judge abnormal situations in advance before the gas limit threshold, ensure equipment safety and operation and maintenance personnel safety, and respond quickly.
Smart Images

Figure CN118707050B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of power distribution room dynamic environment monitoring, and in particular to a power distribution room gas monitoring method and system. Background Art
[0002] The distribution room is usually located inside or near a building and is used to receive, convert, distribute and transmit electrical energy. The distribution room usually consists of a group of high and low voltage distribution devices, transformers, control equipment, and protection devices. The main function of the distribution room is to ensure the safe, reliable, economical and efficient supply of electrical energy to ensure the normal operation of the power system.
[0003] There are a large number of distribution rooms. According to survey data, the number of distribution rooms in Sichuan Province will reach tens of thousands in 2023. At the same time, they are widely distributed and dispersed. Furthermore, most of the distribution rooms currently in operation are located in basements, with a closed operating environment, poor ventilation, and relatively high temperature and humidity. Especially in summer, the temperature is high and the precipitation is large. The distribution room is in a high temperature and humid working environment for a long time. In this case, it is very easy for equipment to become moldy, insulation to decrease, high temperature overheating, and frequent partial discharge to occur, which seriously affects the normal operation of the distribution room. Traditional distribution room maintenance work includes daily inspections, periodic maintenance, and fault maintenance. The existing problems include heavy workload, poor predictability, and frequent over-maintenance and under-maintenance of equipment. In order to improve the efficiency and effectiveness of supervision of the working conditions of the distribution room, realize automatic control of the environment in the distribution room, and ensure the personal safety of the corresponding personnel and the property safety of the distribution room, the existing technology provides dynamic environment monitoring products / systems for distribution rooms. Regarding the gas sensors used in the power distribution room, the main gas components they monitor include: oxygen, ozone, sulfur hexafluoride and hydrogen sulfide. Functionally, the oxygen concentration can be used to reflect the oxidation of the power distribution equipment and the fire accidents in the power distribution room; the ozone concentration can reflect the partial discharge of the power distribution room; the sulfur hexafluoride concentration can reflect the sealing of the insulating gas; the hydrogen sulfide concentration can reflect whether the capacitor in the capacitor cabinet has burned or exploded. At the same time, the above gas concentration detection also has a personal safety protection effect on the personnel entering the power distribution room.
[0004] In summary, the detection results of gas sensors can be used to judge abnormal conditions in distribution rooms. In order to optimize the performance of dynamic environment monitoring products / systems, it is necessary to further optimize related gas monitoring technologies. Summary of the invention
[0005] In response to the above-mentioned problem of further optimizing the relevant gas monitoring technology, the present invention provides a gas monitoring method and system for a distribution room. This solution can be used to judge abnormal conditions in the distribution room in advance and has predictive power for possible abnormal conditions.
[0006] In view of the above problems, a gas monitoring method and system for a power distribution room provided by the present invention solves the problems through the following technical points: a gas monitoring method for a power distribution room, the method monitors the gas phase concentration of a target gas in the power distribution room based on a gas sensor, the gas sensor includes at least one of an oxygen concentration sensor and a sulfur hexafluoride concentration sensor, in the method, a collection value of the gas sensor is collected and the environmental parameters at the time of collection of the collection value are obtained, the environmental parameters include temperature parameters and humidity parameters in the power distribution room;
[0007] The collected values and environmental parameters are input into a gas phase concentration prediction model. The gas phase concentration prediction model predicts the change law of the gas phase concentration of the target gas in the power distribution room within a preset time period according to the collected values and environmental parameters, and obtains a relationship curve between the gas phase concentration of the target gas and time within the preset time period. In the relationship curve, the time when the collected values are collected is used as the initial time, and the collected values are used as the gas phase concentration at the initial time.
[0008] During the preset time period, according to the set gas sensor data collection rules, the gas phase concentration of the target gas in the power distribution room is monitored to obtain the actual concentration value of the target gas;
[0009] In the relationship curve, the corresponding predicted gas phase concentration value is searched according to the actual concentration value acquisition time, and the actual concentration value is compared with the predicted gas phase concentration value. When the comparison result deviates from the set deviation threshold, it is determined that the gas concentration in the power distribution room is abnormal;
[0010] Wherein, during the preset time period, the ventilation and access control conditions of the power distribution room are monitored. When it is detected that the access control is opened or the ventilation system is started, the collected actual concentration value is compared with the preset target gas limit concentration threshold. When the comparison result is that the actual concentration value exceeds the target gas limit concentration threshold, it is determined that the gas concentration in the power distribution room is abnormal;
[0011] When it is detected that the door access control is closed and the ventilation system is closed, the collected values and environmental parameters are collected again, and a new relationship curve is obtained through the gas phase concentration prediction model.
[0012] In the prior art, dynamic environment products are used to realize intelligent monitoring of the operation and environmental conditions of the distribution room, wherein the environmental conditions include indoor air composition, temperature and humidity, etc. The related front-end monitoring equipment includes gas sensors and microenvironment sensors, wherein the gas sensor is used to detect the gas composition, and the microenvironment sensor is used to detect the temperature and humidity. The detected gas components include sulfur hexafluoride, oxygen, ozone and hydrogen sulfide, etc. At the same time, in the prior art, such as the technical solution provided by patent application number CN201921711622.X, for the target gas in the distribution room, when the content is abnormal, the deposited sulfur hexafluoride gas is extracted by the fan and the air is renewed by fresh air. In other technical solutions provided by patent application number CN202311774157.5, the concentration of the target gas is monitored and compared with the threshold. When the threshold is exceeded, the subsequent action signal is triggered. Therefore, the conventional method for processing the target gas in the distribution room is generally as follows: setting thresholds for different target gases, and when the actual concentration of the target gas is monitored to exceed the set threshold, triggering one or more of the following: local alarm, linkage fan, abnormal remote transmission, etc. For example, when the oxygen concentration is lower than 18%, the ozone concentration is higher than 0.2ppm, the hydrogen sulfide concentration is higher than 10ppm, and the sulfur hexafluoride gas concentration is greater than 1000ppm, the relevant signal is triggered. Taking the signal linkage fan start as an example, when the oxygen concentration reaches 20.8%, the ozone concentration, hydrogen sulfide concentration, and sulfur hexafluoride gas concentration are 0, the fan is turned off.
[0013] Effective monitoring of gas in the distribution room is an important method to understand the equipment status in the distribution room, ensure the personal safety of operation and maintenance personnel, and discover equipment accidents. In view of the gas monitoring method in the existing technology that uses thresholds as limit values, this solution provides a technical solution that can predict abnormal conditions that may occur in the distribution room.
[0014] Specifically, in order to meet the requirements of dust prevention and rodent prevention, the distribution room is generally in a closed state with the access control closed during use. At the same time, the ventilation conditions of the distribution room include natural ventilation and forced mechanical ventilation. During the use of the distribution room, due to load changes, weather reasons, etc., there are temperature fluctuations and humidity fluctuations in the room. For the monitoring of oxygen content in the distribution room, firstly, when the temperature rises, the air density decreases relatively, and the oxygen content per unit volume of air decreases. When the air humidity increases, the water vapor component in the air increases accordingly, which makes the oxygen content in the air relatively lower. Secondly, although the electrochemical oxygen sensor uses a temperature compensation coefficient to compensate for the detection results, there is still a problem of differences in measurement errors at different temperatures. Finally, when the distribution room is completely closed or the natural ventilation volume remains unchanged, the metal components in the distribution room undergo oxidation reaction with oxygen, which will cause the detected oxygen concentration to be slightly reduced for a long time. , and the temperature is related to the rate of the oxidation reaction. For the monitoring of sulfur hexafluoride content, if the arc-extinguishing gas leakage from the transformer, high-voltage switch, and circuit breaker cannot be eliminated in time, the detection result of sulfur hexafluoride will continue to increase. At the same time, for the commonly used infrared gas sensors, when the humidity increases, the scattering and absorption of infrared radiation by water vapor are significantly enhanced, and the temperature will change the performance parameters of its electronic components. Therefore, the applicant believes that: even for the same gas phase composition in the distribution room, when the environmental parameters in the distribution room are different, the oxygen concentration sensor and the sulfur hexafluoride concentration sensor will have different detection results. If it is only based on the threshold value as the lower limit value of the oxygen concentration and the threshold value as the upper limit value of the sulfur hexafluoride concentration to judge whether the gas phase concentration of the corresponding target gas in the distribution room is abnormal, the subsequent action signal will be triggered only when the gas phase concentration of the target gas reaches the threshold value, and the abnormal gas situation in the distribution room cannot be effectively judged in advance.
[0015] When this scheme is used in practice, the gas sensor is installed in the distribution room, and a corresponding microenvironment sensor is installed in the distribution room to detect the environmental parameters when the gas sensor obtains the collected value. The gas sensor and the microenvironment sensor both communicate with a data processing module equipped with a gas phase concentration prediction model to input the collected value and the environmental parameters into the gas phase concentration prediction model. The gas phase concentration prediction model predicts the change law of the gas phase concentration of the target gas within a preset time period after the collection time of the collected value based on the current collected value and the environmental parameter, and obtains the relationship curve. The target gas is oxygen and / or sulfur hexafluoride. When the collection time of the subsequent collected value is within the preset time period, for the actual concentration value detected, the predicted gas phase concentration value at the time of obtaining the actual concentration value found from the relationship curve is compared with the actual concentration value, and when the comparison result deviates from the set deviation threshold, it is determined that the gas concentration in the distribution room is abnormal.
[0016] In the above process, the collected values and environmental parameters are used to predict the relationship curve within the preset time period, that is, the environmental parameters are taken as the consideration factors of the relationship curve. In this way, the relatively consistent gas sensor accuracy under the same or similar environmental parameters, the relatively consistent oxygen consumption rate of the oxidation reaction, the relatively consistent leakage rate of the arc extinguishing gas, the relatively consistent target gas content per unit volume and other characteristics are utilized, so that within the preset time period, the normal target gas gas phase concentration at each moment can be predicted and reflected by the relationship curve, and then, according to the comparison result between the actual concentration value and the predicted gas phase concentration value, it is determined whether the gas concentration in the distribution room is abnormal. Therefore, this scheme is adopted to take the environmental parameters as the basis for obtaining the relationship curve, that is, the obtained relationship curve is when In this way, the difference in predicted gas phase concentration values caused by various factors under different environmental parameters is avoided, so that the predicted gas phase concentration value in the relationship curve has higher accuracy. Through the corresponding comparison results, the abnormal change of the gas phase concentration of the target gas in the preset time period can be effectively monitored. For example, before sulfur hexafluoride leaks and deposits to the preset sulfur hexafluoride limit concentration threshold, it can be determined that the sulfur hexafluoride leakage is abnormal, and it can be used to judge the abnormal leakage of sulfur hexafluoride; before the oxygen concentration decreases to the preset oxygen limit concentration threshold, it can be determined whether the oxygen consumption is abnormal, and it can be judged by the intensified oxidation of metal components and the combustion in the room, so that this scheme can be used to judge the abnormal situation of the distribution room in advance, and it has predictive power for possible abnormal situations.
[0017] At the same time, in the preset time period, after obtaining the relationship curve, the discrimination result can be obtained by comparing the actual concentration value detected later with the corresponding predicted gas phase concentration value, so the present solution also has the characteristic of fast response speed.
[0018] In specific implementation, the shorter the preset time period is, the more reliable the monitoring effect is; the higher the collection frequency in the data collection rule is, the more reliable the monitoring effect is. Technical personnel in this field can set it as needed in specific application. For example, the duration of the preset time period is 1 hour, specifically when the initial time is M, the preset time period covers M to M+1; as in the data collection rule, when the deviation between the collection value collected this time and the collection value collected last time is within the set range, the gas sensor adopts a collection value sampling frequency of times / 10min, and when it exceeds the set range, a collection value sampling frequency of times / 1min is adopted.
[0019] The changes in ventilation conditions and access control conditions monitored during the preset time period are considered to be changes in environmental parameters and gas phase components after ventilation and door opening. Therefore, the traditional target gas limit concentration threshold is used to determine whether the gas concentration in the distribution room is abnormal. When the access control is closed and the ventilation system is turned off, it is determined that the distribution room has stable gas phase components and environmental parameters. At this time, the relationship curve obtained by the gas phase concentration prediction model and the actual concentration value measured in real time are used to determine whether there is an abnormal gas concentration.
[0020] In specific implementation, the gas phase concentration prediction model can be established manually or through computer training. Since different distribution rooms have different equipment settings and layouts, different geographical locations, environmental conditions and working parameters, and there are repairs and maintenance during the use cycle of the distribution room, the better way is to establish a gas phase concentration prediction model through the historical usage data of the distribution room. For example, during the historical use cycle of the distribution room, the historical usage data collected under normal use are statistically analyzed, and the change law of the gas phase concentration of the target gas under different environmental parameters is statistically analyzed from these historical usage data. When predicting the relationship curve, the change law corresponding to the environmental parameter in the database is obtained through the current environmental parameter, and the collection time of the collection value is used as the initial time, and the collection value is used as the gas phase concentration at the initial time to obtain the relationship curve.
[0021] Since there is natural ventilation in the distribution room, but the natural ventilation volume may change due to external environmental factors, this introduces interference factors for abnormal judgment based on the relationship curve. A specific method of using this solution can be: during the use cycle of the distribution room, all ventilation in the distribution room is closed for a short period of time (such as 1 hour), and the relationship curve is used to make abnormal judgments. The relationship curve predicted by the corresponding gas phase concentration prediction model should also be under the same distribution room ventilation conditions. In order to make the highest temperature of the distribution room in the short term within the allowable range, it is preferred that the short term is a time period at night when the load of the distribution room is relatively small.
[0022] As a further technical solution of the gas monitoring method in the power distribution room:
[0023] The method for the gas phase concentration prediction model to obtain the relationship curve is: according to the input collection value and environmental parameters, a sub-model closest to the current environmental parameters is searched from multiple sub-models in the database, different sub-models are configured with different environmental parameters, and the sub-model has the change law of the gas phase concentration of the target gas under the configured environmental parameters;
[0024] The collected value is used as the gas phase concentration at the initial moment of the relationship curve, and the gas phase concentration at other moments in the preset time period is generated according to the change law in the sub-model found;
[0025] The relationship curve is obtained by curve fitting according to the gas phase concentration at the initial moment and the gas phase concentration at other moments. This solution is: for the configuration differences, usage environment differences and working parameter differences of different distribution rooms, since the gas phase concentration prediction model established by manual experience assignment is not representative and accurate, and the method of obtaining the corresponding model by computer automatic training does not have a sufficient data basis and requires a lot of computing power, the better application is to obtain a sub-model based on the historical usage data of each distribution room as a data set, and each sub-model is configured with environmental parameters and the change law of the gas phase concentration of the target gas under the environmental parameters. When obtaining the relationship curve, first, the sub-model whose configured environmental parameters are closest to the current environmental parameters is found through the current environmental parameters, and then the change law in the sub-model is used (it can be a fitting of the change trend of the gas phase concentration of the target gas over time in the sub-model from multiple historical usage data, and the trend is the change law), the current collection value is used as the gas phase concentration at the initial moment in the relationship curve, and the change law is used to calculate the predicted gas phase concentration values corresponding to some other moments extracted from the preset time period, and finally, the gas phase concentration values at the initial moment and other moments are used to obtain the relationship curve by curve fitting. In specific implementation, the environmental parameters that determine the sub-model screening include temperature parameters and humidity parameters. Since the current environmental parameters are not necessarily reproducible in the historical usage data, in the process of completing the sub-model screening, technical personnel in this field can configure relevant weight factors for the temperature parameters and humidity parameters. When the environmental parameters configured in a sub-model are consistent with the current environmental parameters, the sub-model is selected as the sub-model with the configuration change rule. When the environmental parameters consistent with the current environmental parameters cannot be matched in the historical usage data, the sub-model is screened using the weight factor calculation results. The specific method can be: from all sub-models, according to the temperature parameters and environmental parameters configured for each sub-model, multiple sub-models are screened out, and the temperature parameters or humidity parameters configured in these sub-models are close to the current environmental parameters. The difference between the temperature and humidity parameters in these sub-models and the relevant parameters in the current environmental parameters is obtained, and the optimal sub-model is further screened out under the consideration of the corresponding weight factors.
[0026] The sub-model is obtained through the historical usage data of the current power distribution room: the historical environmental parameters of the power distribution room are divided into multiple parameter intervals according to the value size, and the historical usage data monitored during the continuous use cycle without ventilation and access control is searched and the gas phase concentration change law of each parameter interval is assigned;
[0027] The method of assigning values is as follows: for any parameter interval, multiple historical target gas gas phase concentrations are collected from historical usage data, and the environmental parameters at the time of collecting each historical target gas gas phase concentration are within the parameter interval, and the change law of the target gas gas phase concentration over time in the parameter interval is calculated by using the historical target gas gas phase concentrations;
[0028] After the power distribution room is overhauled, the sub-model is reconfigured;
[0029] The parameter interval is a temperature parameter interval divided according to the temperature parameter, and different parameter intervals have the same humidity parameter range. In the process of searching the sub-model, the sub-model is searched according to the temperature parameter. When the humidity parameter input to the gas phase concentration prediction model belongs to the outside of the humidity parameter range, the gas phase concentration prediction model triggers the power distribution room to adjust the humidity;
[0030] Different sub-models are configured with different temperature parameter ranges.
[0031] This scheme is: the environmental parameters in the historical usage data include a large interval range, and these ranges are divided into multiple parameter intervals as small intervals. The continuous use period is the period in which the indoor gas phase environment does not fluctuate greatly under the influence of the outside world, and the historical usage data within the period is used to assign values to ensure the value of the historical usage number. Further, the gas phase concentrations of multiple historical target gases under the parameter interval are used to extract and count the change rules. After the power distribution room is overhauled, the oxygen consumption of the oxygen consumption point and the leakage of the sulfur hexafluoride leakage point may change, so the sub-model is rebuilt. Considering the possible arc flash hazards, the power distribution room is generally equipped with a dehumidifier, so the humidity change interval allowed in the power distribution room is set to be smaller than the temperature change interval allowed. Therefore, when dividing the parameter interval, only the humidity parameter is classified into normal humidity and abnormal humidity. The humidity parameter range of the parameter interval is the normal humidity range. A specific application is to search only according to the temperature parameter when searching for the sub-model. When the humidity parameter input to the gas phase concentration prediction model is abnormal humidity, the method triggers humidity adjustment.
[0032] The sub-model is updated during the use of the power distribution room, and the updating method is: within the preset time period, when the comparison result of the actual concentration value of the target gas in the power distribution room at each moment and the corresponding predicted gas phase concentration value in the relationship curve is within the set deviation threshold, and the door opening and ventilation system startup are not detected within the preset time period, according to the change law of the actual concentration value within the preset time period, the change law in the sub-model found is updated. The above scheme is: because there is a development law for the consumption of oxygen by metal components and the leakage of arc extinguishing gas, and the performance attenuation of gas sensors also has a development law, when the actual concentration value collected during the use of the power distribution room is determined to be normal use data, the collected value is included in the sub-model update, so that the subsequent collected values have the latest sub-model when performing abnormal comparison to reduce the probability of misjudgment of gas abnormalities. The sub-model found is the sub-model used for abnormal comparison of actual concentration values. The failure to detect the opening of the door and the startup of the ventilation system means that these actual concentration values have normal data characteristics within the preset time period, which is used to ensure the value of the data used for updating.
[0033] The deviation threshold includes a first deviation threshold and a second deviation threshold, the first deviation threshold is used to compare the actual oxygen concentration value with the predicted oxygen concentration value, and the second deviation threshold is used to compare the actual sulfur hexafluoride concentration value with the predicted sulfur hexafluoride concentration value;
[0034] For the actual oxygen concentration value collected, within a preset time period, when the actual concentration value currently collected is greater than the actual concentration value collected last time, the collected value and the environmental parameters are collected again, and a new relationship curve is obtained through the gas phase concentration prediction model; when the actual concentration value currently collected is less than the actual concentration value collected last time, the actual concentration value is compared with the predicted gas phase concentration value using the first deviation threshold;
[0035] For the actual concentration of sulfur hexafluoride collected, within a preset time period, when the current actual concentration value collected is less than the actual concentration value collected last time, the collected value and environmental parameters are recollected, and a new relationship curve is obtained through the gas phase concentration prediction model; when the current actual concentration value collected is greater than the actual concentration value collected last time, the second deviation threshold is used to compare the actual concentration value with the predicted gas phase concentration value. This solution is: for oxygen, when it is determined that the actual concentration value currently collected is greater than the actual concentration value collected last time, it indicates that the natural ventilation situation has changed and caused the oxygen concentration to increase. At this time, a new relationship curve should be established to match the current environmental parameters. When the actual concentration value currently collected is less than the actual concentration value collected last time, it indicates that the natural ventilation situation has little effect on the current environmental parameters and the gas phase concentration, and the first deviation threshold is used for gas anomaly judgment; different from the oxygen anomaly judgment, when the actual concentration value currently collected is less than the actual concentration value collected last time, it indicates that natural ventilation has accelerated the discharge of sulfur hexafluoride. At this time, natural ventilation also has a greater impact on the gas phase composition and environmental parameters in the room. In order to improve the reliability of gas anomaly monitoring, the relationship curve should be reconfigured at this time.
[0036] The oxygen concentration sensor is an electrochemical gas sensor;
[0037] The number of oxygen concentration sensors is greater than or equal to 2, and different oxygen concentration sensors are set at different locations in the power distribution room. The gas phase concentration prediction model configures a relationship curve for each oxygen concentration sensor. When the actual concentration value collected by any oxygen concentration sensor is compared to the abnormal gas concentration in the power distribution room, it is determined that the gas concentration in the power distribution room is abnormal. The above provides a specific monitoring method for oxygen concentration. For oxygen, due to the relative position of the oxygen consumption point and the sensor in the room and the airflow model in the room, different locations may have different oxygen concentration measurement results and change rules. The above method can effectively improve the globality of oxygen concentration monitoring in the power distribution room.
[0038] The sulfur hexafluoride concentration sensor is an infrared gas sensor, and the method for obtaining the collected value of the sulfur hexafluoride concentration sensor is:
[0039] S1. Determine the location of the leakage point where sulfur hexafluoride may leak in the power distribution room;
[0040] S2. According to the distribution of the leakage points in the power distribution room, the sulfur hexafluoride deposition area located below the leakage point is marked in the power distribution room;
[0041] S3. The sulfur hexafluoride concentration sensor obtains the actual concentration value of sulfur hexafluoride from the sulfur hexafluoride deposition area according to the set data collection rules. The above provides a specific monitoring method for sulfur hexafluoride concentration, that is, to identify the deposition area through possible leak points, and then conduct targeted data collection on the deposition area to improve the reliability of sulfur hexafluoride gas concentration monitoring.
[0042] When the number of the sulfur hexafluoride deposition areas is greater than or equal to 2, each sulfur hexafluoride deposition area is equipped with a sulfur hexafluoride concentration sensor;
[0043] The gas phase concentration prediction model configures a relationship curve for each sulfur hexafluoride deposition area. When the actual concentration value collected from any sulfur hexafluoride precipitation area is compared to the abnormal gas concentration in the distribution room, it is determined that the gas concentration in the distribution room is abnormal. The above provides a sulfur hexafluoride monitoring method corresponding to multiple deposition areas. Due to the relative position of the leakage position and the sensor in the room and the airflow model in the room, different positions may have different sulfur hexafluoride concentration measurement results and change rules. The above method can effectively improve the overall nature of sulfur hexafluoride concentration monitoring in the distribution room.
[0044] The present solution also relates to a gas monitoring system for a power distribution room, which is used to implement any of the methods described above;
[0045] The system includes a gas sensor and a temperature and humidity sensor, wherein the gas sensor includes at least one of an oxygen concentration sensor and a sulfur hexafluoride concentration sensor;
[0046] The system includes a data processing module, in which a gas phase concentration prediction model is provided, and the gas sensor and the temperature and humidity sensor are both data-connected to the data processing module. The gas phase concentration prediction model predicts the change law of the gas phase concentration of the target gas in the power distribution room within a preset time period according to the collected values of the gas sensor and the environmental parameters collected by the temperature and humidity sensor, and obtains a relationship curve between the gas phase concentration of the target gas and time within the preset time period, in which the collection time of the collected value is used as the initial time, and the collected value is used as the gas phase concentration at the initial time;
[0047] During the preset time period, the gas sensor monitors the gas phase concentration of the target gas in the power distribution room according to the set data collection rules to obtain the actual concentration value of the target gas;
[0048] In the relationship curve, the gas phase concentration prediction model searches for the corresponding predicted gas phase concentration value according to the actual concentration value acquisition time, and compares the actual concentration value with the predicted gas phase concentration value. When the comparison result deviates from the set deviation threshold, it is determined that the gas concentration in the power distribution room is abnormal;
[0049] Among them, in the preset time period, the data processing module monitors the ventilation and access control conditions of the power distribution room. When it is detected that the access control is opened or the ventilation system is started, the collected actual concentration value is compared with the preset target gas limit concentration threshold. When the comparison result is that the actual concentration value exceeds the target gas limit concentration threshold, it is determined that the gas concentration in the power distribution room is abnormal;
[0050] When the data processing module detects that the door access is closed and the ventilation system is closed, the collected values and environmental parameters are collected again, and a new relationship curve is obtained through the gas phase concentration prediction model. It is easy to understand that this system is used to implement the above method.
[0051] As a further technical solution for the gas monitoring system in the power distribution room:
[0052] The oxygen concentration sensor is an electrochemical gas sensor, and the number of the oxygen concentration sensors is greater than or equal to 2;
[0053] The sulfur hexafluoride concentration sensor is an infrared gas sensor including an air chamber, and the air chamber is provided with an air inlet and an air outlet, and the air inlet and the air outlet can be switched with each other. In the above scheme, the air chamber is a closed place for the infrared gas sensor to detect the sulfur hexafluoride concentration, and the air inlet and the air outlet can be switched with each other, which is a configuration of limiting the air chamber. With this scheme, the air inlet and the air outlet can be switched to change the air inlet position of the air chamber, so that the infrared gas sensor can monitor the sulfur hexafluoride gas concentration collected at different directions around it at its installation position.
[0054] The present invention has the following beneficial effects:
[0055] In this scheme, environmental parameters are used as the basis for obtaining the relationship curve, which avoids the differences in predicted gas phase concentration values caused by various factors under different environmental parameters, so that the predicted gas phase concentration values in the relationship curve have higher accuracy. Through the corresponding comparison results, the abnormal changes in the gas phase concentration of the target gas within the preset time period can be effectively monitored. Compared with determining whether the gas concentration is abnormal based only on the comparison of the target gas limit concentration threshold, this scheme can be used to judge abnormal conditions in the distribution room in advance and has predictive power for possible abnormal conditions.
[0056] In this solution, in the preset time period, after obtaining the relationship curve, the discrimination result can be obtained by comparing the actual concentration value detected later with the corresponding predicted gas phase concentration value, so this solution also has the characteristic of fast response speed. BRIEF DESCRIPTION OF THE DRAWINGS
[0057] Figure 1 The present invention is a flowchart of a specific embodiment of a method for monitoring gas in a power distribution room described in the present invention. DETAILED DESCRIPTION
[0058] The present invention is further described in detail below in conjunction with embodiments, but the present invention is not limited to the following embodiments:
[0059] Embodiment 1:
[0060] Combination Figure 1 , a gas monitoring method for a power distribution room, the method monitors the gas phase concentration of a target gas in the power distribution room based on a gas sensor, the gas sensor includes at least one of an oxygen concentration sensor and a sulfur hexafluoride concentration sensor, in the method, a collection value of the gas sensor is collected and an environmental parameter at the time of collecting the collection value is obtained, the environmental parameter includes a temperature parameter and a humidity parameter in the power distribution room;
[0061] The collected values and environmental parameters are input into a gas phase concentration prediction model. The gas phase concentration prediction model predicts the change law of the gas phase concentration of the target gas in the power distribution room within a preset time period according to the collected values and environmental parameters, and obtains a relationship curve between the gas phase concentration of the target gas and time within the preset time period. In the relationship curve, the time when the collected values are collected is used as the initial time, and the collected values are used as the gas phase concentration at the initial time.
[0062] During the preset time period, according to the set gas sensor data collection rules, the gas phase concentration of the target gas in the power distribution room is monitored to obtain the actual concentration value of the target gas;
[0063] In the relationship curve, the corresponding predicted gas phase concentration value is searched according to the actual concentration value acquisition time, and the actual concentration value is compared with the predicted gas phase concentration value. When the comparison result deviates from the set deviation threshold, it is determined that the gas concentration in the power distribution room is abnormal;
[0064] Wherein, during the preset time period, the ventilation and access control conditions of the power distribution room are monitored. When it is detected that the access control is opened or the ventilation system is started, the collected actual concentration value is compared with the preset target gas limit concentration threshold. When the comparison result is that the actual concentration value exceeds the target gas limit concentration threshold, it is determined that the gas concentration in the power distribution room is abnormal;
[0065] When it is detected that the door access control is closed and the ventilation system is closed, the collected values and environmental parameters are collected again, and a new relationship curve is obtained through the gas phase concentration prediction model.
[0066] In the prior art, dynamic environment products are used to realize intelligent monitoring of the operation and environmental conditions of the distribution room, wherein the environmental conditions include indoor air composition, temperature and humidity, etc. The related front-end monitoring equipment includes gas sensors and microenvironment sensors, wherein the gas sensor is used to detect the gas composition, and the microenvironment sensor is used to detect the temperature and humidity. The detected gas components include sulfur hexafluoride, oxygen, ozone and hydrogen sulfide, etc. At the same time, in the prior art, such as the technical solution provided by patent application number CN201921711622.X, for the target gas in the distribution room, when the content is abnormal, the deposited sulfur hexafluoride gas is extracted by the fan and the air is renewed by fresh air. In other technical solutions provided by patent application number CN202311774157.5, the concentration of the target gas is monitored and compared with the threshold. When the threshold is exceeded, the subsequent action signal is triggered. Therefore, the conventional method for processing the target gas in the distribution room is generally as follows: setting thresholds for different target gases, and when the actual concentration of the target gas is monitored to exceed the set threshold, triggering one or more of the following: local alarm, linkage fan, abnormal remote transmission, etc. For example, when the oxygen concentration is lower than 18%, the ozone concentration is higher than 0.2ppm, the hydrogen sulfide concentration is higher than 10ppm, and the sulfur hexafluoride gas concentration is greater than 1000ppm, the relevant signal is triggered. Taking the signal linkage fan start as an example, when the oxygen concentration reaches 20.8%, the ozone concentration, hydrogen sulfide concentration, and sulfur hexafluoride gas concentration are 0, the fan is turned off.
[0067] Effective monitoring of gas in the distribution room is an important method to understand the equipment status in the distribution room, ensure the personal safety of operation and maintenance personnel, and discover equipment accidents. In view of the gas monitoring method in the existing technology that uses thresholds as limit values, this solution provides a technical solution that can predict abnormal conditions that may occur in the distribution room.
[0068] Specifically, in order to meet the requirements of dust prevention and rodent prevention, the distribution room is generally in a closed state with the access control closed during use. At the same time, the ventilation conditions of the distribution room include natural ventilation and forced mechanical ventilation. During the use of the distribution room, due to load changes, weather reasons, etc., there are temperature fluctuations and humidity fluctuations in the room. For the monitoring of oxygen content in the distribution room, firstly, when the temperature rises, the air density decreases relatively, and the oxygen content per unit volume of air decreases. When the air humidity increases, the water vapor component in the air increases accordingly, which makes the oxygen content in the air relatively lower. Secondly, although the electrochemical oxygen sensor uses a temperature compensation coefficient to compensate for the detection results, there is still a problem of differences in measurement errors at different temperatures. Finally, when the distribution room is completely closed or the natural ventilation volume remains unchanged, the metal components in the distribution room undergo oxidation reaction with oxygen, which will cause the detected oxygen concentration to be slightly reduced for a long time. , and the temperature is related to the rate of the oxidation reaction. For the monitoring of sulfur hexafluoride content, if the arc-extinguishing gas leakage from the transformer, high-voltage switch, and circuit breaker cannot be eliminated in time, the detection result of sulfur hexafluoride will continue to increase. At the same time, for the commonly used infrared gas sensors, when the humidity increases, the scattering and absorption of infrared radiation by water vapor are significantly enhanced, and the temperature will change the performance parameters of its electronic components. Therefore, the applicant believes that: even for the same gas phase composition in the distribution room, when the environmental parameters in the distribution room are different, the oxygen concentration sensor and the sulfur hexafluoride concentration sensor will have different detection results. If it is only based on the threshold value as the lower limit value of the oxygen concentration and the threshold value as the upper limit value of the sulfur hexafluoride concentration to judge whether the gas phase concentration of the corresponding target gas in the distribution room is abnormal, the subsequent action signal will be triggered only when the gas phase concentration of the target gas reaches the threshold value, and the abnormal gas situation in the distribution room cannot be effectively judged in advance.
[0069] When this scheme is used in practice, the gas sensor is installed in the distribution room, and a corresponding microenvironment sensor is installed in the distribution room to detect the environmental parameters when the gas sensor obtains the collected values. The gas sensor and the microenvironment sensor both communicate with the gas phase concentration prediction model to input the collected values and the environmental parameters into the gas phase concentration prediction model. The gas phase concentration prediction model predicts the change law of the gas phase concentration of the target gas within a preset time period after the collection time of the collected values based on the current collected values and the environmental parameters, and obtains the relationship curve. The target gas is oxygen and / or sulfur hexafluoride. When the collection time of the subsequent collected values is within the preset time period, for the actual concentration value detected, the predicted gas phase concentration value at the time of obtaining the actual concentration value found from the relationship curve is compared with the actual concentration value, and when the comparison result deviates from the set deviation threshold, it is determined that the gas concentration in the distribution room is abnormal.
[0070] In the above process, the collected values and environmental parameters are used to predict the relationship curve within the preset time period, that is, the environmental parameters are taken as the consideration factors of the relationship curve. In this way, the relatively consistent gas sensor accuracy under the same or similar environmental parameters, the relatively consistent oxygen consumption rate of the oxidation reaction, the relatively consistent leakage rate of the arc extinguishing gas, the relatively consistent target gas content per unit volume and other characteristics are utilized, so that within the preset time period, the normal target gas gas phase concentration at each moment can be predicted and reflected by the relationship curve, and then, according to the comparison result between the actual concentration value and the predicted gas phase concentration value, it is determined whether the gas concentration in the distribution room is abnormal. Therefore, this scheme is adopted to take the environmental parameters as the basis for obtaining the relationship curve, that is, the obtained relationship curve is when In this way, the difference in predicted gas phase concentration values caused by various factors under different environmental parameters is avoided, so that the predicted gas phase concentration value in the relationship curve has higher accuracy. Through the corresponding comparison results, the abnormal change of the gas phase concentration of the target gas in the preset time period can be effectively monitored. For example, before sulfur hexafluoride leaks and deposits to the preset sulfur hexafluoride limit concentration threshold, it can be determined that the sulfur hexafluoride leakage is abnormal, and it can be used to judge the abnormal leakage of sulfur hexafluoride; before the oxygen concentration decreases to the preset oxygen limit concentration threshold, it can be determined whether the oxygen consumption is abnormal, and it can be judged by the intensified oxidation of metal components and the combustion in the room, so that this scheme can be used to judge the abnormal situation of the distribution room in advance, and it has predictive power for possible abnormal situations.
[0071] At the same time, in the preset time period, after obtaining the relationship curve, the discrimination result can be obtained by comparing the actual concentration value detected later with the corresponding predicted gas phase concentration value, so the present solution also has the characteristic of fast response speed.
[0072] In specific implementation, the shorter the preset time period is, the more reliable the monitoring effect is; the higher the collection frequency in the data collection rule is, the more reliable the monitoring effect is. Technical personnel in this field can set it as needed in specific application. For example, the duration of the preset time period is 1 hour, specifically when the initial time is M, the preset time period covers M to M+1; as in the data collection rule, when the deviation between the collection value collected this time and the collection value collected last time is within the set range, the gas sensor adopts a collection value sampling frequency of times / 10min, and when it exceeds the set range, a collection value sampling frequency of times / 1min is adopted.
[0073] The changes in ventilation conditions and access control conditions monitored during the preset time period are considered to be changes in environmental parameters and gas phase components after ventilation and door opening. Therefore, the traditional target gas limit concentration threshold is used to determine whether the gas concentration in the distribution room is abnormal. When the access control is closed and the ventilation system is turned off, it is determined that the distribution room has stable gas phase components and environmental parameters. At this time, the relationship curve obtained by the gas phase concentration prediction model and the actual concentration value measured in real time are used to determine whether there is an abnormal gas concentration.
[0074] In specific implementation, the gas phase concentration prediction model can be established manually or through computer training. Since different distribution rooms have different equipment settings and layouts, different geographical locations, environmental conditions and working parameters, and there are repairs and maintenance during the use cycle of the distribution room, the better way is to establish a gas phase concentration prediction model through the historical usage data of the distribution room. For example, during the historical use cycle of the distribution room, the historical usage data collected under normal use are statistically analyzed, and the change law of the gas phase concentration of the target gas under different environmental parameters is statistically analyzed from these historical usage data. When predicting the relationship curve, the change law corresponding to the environmental parameter in the database is obtained through the current environmental parameter, and the collection time of the collection value is used as the initial time, and the collection value is used as the gas phase concentration at the initial time to obtain the relationship curve.
[0075] Since there is natural ventilation in the distribution room, but the natural ventilation volume may change due to external environmental factors, this introduces interference factors for abnormal judgment based on the relationship curve. A specific method of using this solution can be: during the use cycle of the distribution room, all ventilation in the distribution room is closed for a short period of time (such as 1 hour), and the relationship curve is used to make abnormal judgments. The relationship curve predicted by the corresponding gas phase concentration prediction model should also be under the same distribution room ventilation conditions. In order to make the highest temperature of the distribution room in the short term within the allowable range, it is preferred that the short term is a time period at night when the load of the distribution room is relatively small.
[0076] Embodiment 2:
[0077] This embodiment is further refined on the basis of embodiment 1:
[0078] The method for the gas phase concentration prediction model to obtain the relationship curve is: according to the input collection value and environmental parameters, a sub-model closest to the current environmental parameters is searched from multiple sub-models in the database, different sub-models are configured with different environmental parameters, and the sub-model has the change law of the gas phase concentration of the target gas under the configured environmental parameters;
[0079] The collected value is used as the gas phase concentration at the initial moment of the relationship curve, and the gas phase concentration at other moments in the preset time period is generated according to the change law in the sub-model found;
[0080] The relationship curve is obtained by curve fitting according to the gas phase concentration at the initial moment and the gas phase concentration at other moments. This solution is: for the configuration differences, usage environment differences and working parameter differences of different distribution rooms, since the gas phase concentration prediction model established by manual experience assignment is not representative and accurate, and the method of obtaining the corresponding model by computer automatic training does not have a sufficient data basis and requires a lot of computing power, the better application is to obtain a sub-model based on the historical usage data of each distribution room as a data set, and each sub-model is configured with environmental parameters and the change law of the gas phase concentration of the target gas under the environmental parameters. When obtaining the relationship curve, first, the sub-model whose configured environmental parameters are closest to the current environmental parameters is found through the current environmental parameters, and then, the change law in the sub-model is used (the sub-model can be fitted from multiple historical usage data, and the change trend of the gas phase concentration of the target gas over time, which is the change law), the current collection value is used as the gas phase concentration at the initial moment in the relationship curve, and the change law is used to calculate the predicted gas phase concentration values corresponding to some other moments extracted from the preset time period, and finally, the gas phase concentration values at the initial moment and other moments are used to obtain the relationship curve by curve fitting. In specific implementation, the environmental parameters that determine the sub-model screening include temperature parameters and humidity parameters. Since the current environmental parameters are not necessarily reproducible in the historical usage data, in the process of completing the sub-model screening, technical personnel in this field can configure relevant weight factors for the temperature parameters and humidity parameters. When the environmental parameters configured in a sub-model are consistent with the current environmental parameters, the sub-model is selected as the sub-model with the configuration change rule. When the environmental parameters consistent with the current environmental parameters cannot be matched in the historical usage data, the sub-model is screened using the weight factor calculation results. The specific method can be: from all sub-models, according to the temperature parameters and environmental parameters configured for each sub-model, multiple sub-models are screened out, and the temperature parameters or humidity parameters configured in these sub-models are close to the current environmental parameters. The difference between the temperature and humidity parameters in these sub-models and the relevant parameters in the current environmental parameters is obtained, and the optimal sub-model is further screened out under the consideration of the corresponding weight factors.
[0081] Embodiment 3:
[0082] This embodiment is further refined on the basis of Embodiment 2:
[0083] The sub-model is obtained through the historical usage data of the current power distribution room: the historical environmental parameters of the power distribution room are divided into multiple parameter intervals according to the value size, and the historical usage data monitored during the continuous use cycle without ventilation and access control is searched and the gas phase concentration change law of each parameter interval is assigned;
[0084] The method of assigning values is as follows: for any parameter interval, multiple historical target gas gas phase concentrations are collected from historical usage data, and the environmental parameters at the time of collecting each historical target gas gas phase concentration are within the parameter interval, and the change law of the target gas gas phase concentration over time in the parameter interval is calculated by using the historical target gas gas phase concentrations;
[0085] After the power distribution room is overhauled, the sub-model is reconfigured;
[0086] The parameter interval is a temperature parameter interval divided according to the temperature parameter, and different parameter intervals have the same humidity parameter range. In the process of searching the sub-model, the sub-model is searched according to the temperature parameter. When the humidity parameter input to the gas phase concentration prediction model belongs to the outside of the humidity parameter range, the gas phase concentration prediction model triggers the power distribution room to adjust the humidity;
[0087] Different sub-models are configured with different temperature parameter ranges.
[0088] This scheme is: the environmental parameters in the historical usage data include a large interval range, and these ranges are divided into multiple parameter intervals as small intervals. The continuous use period is the period in which the indoor gas phase environment does not fluctuate greatly under the influence of the outside world, and the historical usage data within the period is used to assign values to ensure the value of the historical usage number. Further, the gas phase concentrations of multiple historical target gases under the parameter interval are used to extract and count the change rules. After the power distribution room is overhauled, the oxygen consumption of the oxygen consumption point and the leakage of the sulfur hexafluoride leakage point may change, so the sub-model is rebuilt. Considering the possible arc flash hazards, the power distribution room is generally equipped with a dehumidifier, so the humidity change interval allowed in the power distribution room is set to be smaller than the temperature change interval allowed. Therefore, when dividing the parameter interval, only the humidity parameter is classified into normal humidity and abnormal humidity. The humidity parameter range of the parameter interval is the normal humidity range. A specific application is to search only according to the temperature parameter when searching for the sub-model. When the humidity parameter input to the gas phase concentration prediction model is abnormal humidity, the method triggers humidity adjustment.
[0089] Embodiment 4:
[0090] This embodiment is further refined on the basis of Embodiment 2:
[0091] The sub-model is updated during the use of the power distribution room, and the updating method is: within the preset time period, when the comparison result of the actual concentration value of the target gas in the power distribution room at each moment and the corresponding predicted gas phase concentration value in the relationship curve is within the set deviation threshold, and the door opening and ventilation system startup are not detected within the preset time period, according to the change law of the actual concentration value within the preset time period, the change law in the sub-model found is updated. The above scheme is: because there is a development law for the consumption of oxygen by metal components and the leakage of arc extinguishing gas, and the performance attenuation of gas sensors also has a development law, when the actual concentration value collected during the use of the power distribution room is determined to be normal use data, the collected value is included in the sub-model update, so that the subsequent collected values have the latest sub-model when performing abnormal comparison to reduce the probability of misjudgment of gas abnormalities. The sub-model found is the sub-model used for abnormal comparison of actual concentration values. The failure to detect the opening of the door and the startup of the ventilation system means that these actual concentration values have normal data characteristics within the preset time period, which is used to ensure the value of the data used for updating.
[0092] Embodiment 5:
[0093] This embodiment is further refined on the basis of embodiment 1:
[0094] The deviation threshold includes a first deviation threshold and a second deviation threshold, the first deviation threshold is used to compare the actual oxygen concentration value with the predicted oxygen concentration value, and the second deviation threshold is used to compare the actual sulfur hexafluoride concentration value with the predicted sulfur hexafluoride concentration value;
[0095] For the actual oxygen concentration value collected, within a preset time period, when the actual concentration value currently collected is greater than the actual concentration value collected last time, the collected value and the environmental parameters are collected again, and a new relationship curve is obtained through the gas phase concentration prediction model; when the actual concentration value currently collected is less than the actual concentration value collected last time, the actual concentration value is compared with the predicted gas phase concentration value using the first deviation threshold;
[0096] For the actual concentration of sulfur hexafluoride collected, within a preset time period, when the current actual concentration value collected is less than the actual concentration value collected last time, the collected value and environmental parameters are recollected, and a new relationship curve is obtained through the gas phase concentration prediction model; when the current actual concentration value collected is greater than the actual concentration value collected last time, the second deviation threshold is used to compare the actual concentration value with the predicted gas phase concentration value. This solution is: for oxygen, when it is determined that the actual concentration value currently collected is greater than the actual concentration value collected last time, it indicates that the natural ventilation situation has changed and caused the oxygen concentration to increase. At this time, a new relationship curve should be established to match the current environmental parameters. When the actual concentration value currently collected is less than the actual concentration value collected last time, it indicates that the natural ventilation situation has little effect on the current environmental parameters and the gas phase concentration, and the first deviation threshold is used for gas anomaly judgment; different from the oxygen anomaly judgment, when the actual concentration value currently collected is less than the actual concentration value collected last time, it indicates that natural ventilation has accelerated the discharge of sulfur hexafluoride. At this time, natural ventilation also has a greater impact on the gas phase composition and environmental parameters in the room. In order to improve the reliability of gas anomaly monitoring, the relationship curve should be reconfigured at this time.
[0097] Embodiment 6:
[0098] This embodiment is further refined on the basis of embodiment 1:
[0099] The oxygen concentration sensor is an electrochemical gas sensor;
[0100] The number of oxygen concentration sensors is greater than or equal to 2, and different oxygen concentration sensors are set at different locations in the power distribution room. The gas phase concentration prediction model configures a relationship curve for each oxygen concentration sensor. When the actual concentration value collected by any oxygen concentration sensor is compared to the abnormal gas concentration in the power distribution room, it is determined that the gas concentration in the power distribution room is abnormal. The above provides a specific monitoring method for oxygen concentration. For oxygen, due to the relative position of the oxygen consumption point and the sensor in the room and the airflow model in the room, different locations may have different oxygen concentration measurement results and change rules. The above method can effectively improve the globality of oxygen concentration monitoring in the power distribution room.
[0101] Embodiment 7:
[0102] This embodiment is further refined on the basis of embodiment 1:
[0103] The sulfur hexafluoride concentration sensor is an infrared gas sensor, and the method for obtaining the collected value of the sulfur hexafluoride concentration sensor is:
[0104] S1. Determine the location of the leakage point where sulfur hexafluoride may leak in the power distribution room;
[0105] S2. According to the distribution of the leakage points in the power distribution room, the sulfur hexafluoride deposition area located below the leakage point is marked in the power distribution room;
[0106] S3. The sulfur hexafluoride concentration sensor obtains the actual concentration value of sulfur hexafluoride from the sulfur hexafluoride deposition area according to the set data collection rules. The above provides a specific monitoring method for sulfur hexafluoride concentration, that is, to identify the deposition area through possible leak points, and then conduct targeted data collection on the deposition area to improve the reliability of sulfur hexafluoride gas concentration monitoring.
[0107] Embodiment 8:
[0108] This embodiment is further refined on the basis of Embodiment 7:
[0109] When the number of the sulfur hexafluoride deposition areas is greater than or equal to 2, each sulfur hexafluoride deposition area is equipped with a sulfur hexafluoride concentration sensor;
[0110] The gas phase concentration prediction model configures a relationship curve for each sulfur hexafluoride deposition area. When the actual concentration value collected from any sulfur hexafluoride precipitation area is compared to the abnormal gas concentration in the distribution room, it is determined that the gas concentration in the distribution room is abnormal. The above provides a sulfur hexafluoride monitoring method corresponding to multiple deposition areas. Due to the relative position of the leakage position and the sensor in the room and the airflow model in the room, different positions may have different sulfur hexafluoride concentration measurement results and change rules. The above method can effectively improve the overall nature of sulfur hexafluoride concentration monitoring in the distribution room.
[0111] Embodiment 9:
[0112] This embodiment provides a gas monitoring system for a power distribution room based on the embodiment 1, and the system is used to implement the method described in the embodiment 1;
[0113] The system includes a gas sensor and a temperature and humidity sensor, wherein the gas sensor includes at least one of an oxygen concentration sensor and a sulfur hexafluoride concentration sensor;
[0114] The system includes a data processing module, in which a gas phase concentration prediction model is provided, and the gas sensor and the temperature and humidity sensor are both data-connected to the data processing module. The gas phase concentration prediction model predicts the change law of the gas phase concentration of the target gas in the power distribution room within a preset time period according to the collected values of the gas sensor and the environmental parameters collected by the temperature and humidity sensor, and obtains a relationship curve between the gas phase concentration of the target gas and time within the preset time period, in which the collection time of the collected value is used as the initial time, and the collected value is used as the gas phase concentration at the initial time;
[0115] During the preset time period, the gas sensor monitors the gas phase concentration of the target gas in the power distribution room according to the set data collection rules to obtain the actual concentration value of the target gas;
[0116] In the relationship curve, the gas phase concentration prediction model searches for the corresponding predicted gas phase concentration value according to the actual concentration value acquisition time, and compares the actual concentration value with the predicted gas phase concentration value. When the comparison result deviates from the set deviation threshold, it is determined that the gas concentration in the power distribution room is abnormal;
[0117] Among them, in the preset time period, the data processing module monitors the ventilation and access control conditions of the power distribution room. When it is detected that the access control is opened or the ventilation system is started, the collected actual concentration value is compared with the preset target gas limit concentration threshold. When the comparison result is that the actual concentration value exceeds the target gas limit concentration threshold, it is determined that the gas concentration in the power distribution room is abnormal;
[0118] Among them, when the data processing module detects that the access control is closed and the ventilation system is closed, the collected values and environmental parameters are re-collected, and a new relationship curve is obtained through the gas phase concentration prediction model. It is easy to understand that the present system is used to implement the above method. In the specific implementation, the system includes a dynamic ring host, and the dynamic ring host includes a data processing module equipped with a gas phase concentration prediction model. The system also includes an all-in-one gas sensor that communicates with the dynamic ring host. In addition to an oxygen concentration sensor and a sulfur hexafluoride concentration sensor, the all-in-one gas sensor also has an ozone concentration sensor and a hydrogen sulfide concentration sensor. However, since the generation and concentration changes of ozone and hydrogen sulfide are sudden (such as determined by local discharge in the room, combustion or explosion of capacitors in the cabinet), these two gases are not suitable for using a predictive relationship curve to determine whether There are abnormal concentrations. In practice, the dynamic environment host collects the concentration measurement values of the two gases and compares them with the preset concentration thresholds to determine whether the gas concentration exceeds the standard and whether the equipment in the distribution room is working abnormally. At the same time, the partial discharge sensor carried by the system can be used to verify the concentration measurement results of the two gases. At the same time, the system is equipped with a fan linkage device, a microenvironment sensor and a smart door lock. These devices communicate with the dynamic environment host to achieve, for example: monitoring access control through smart door locks, monitoring the operation of the ventilation system through fan linkage devices, detecting the temperature and humidity in the room through microenvironment sensors, and monitoring smoke and water immersion in the room. That is: the data processing module is arranged on the dynamic environment host for monitoring the distribution room, the gas sensor is integrated on the multi-in-one gas sensor, the temperature and humidity sensor is integrated on the micro-environment sensor, and also includes a smart door lock as a component of the distribution room access control, and a fan linkage device for realizing the automatic operation of the distribution room ventilation system. The multi-in-one gas sensor, micro-environment sensor, smart door lock and fan linkage device all communicate with the dynamic environment host, so that the dynamic environment host can obtain the collected values, environmental parameters, access control status and ventilation conditions. In a specific embodiment, the corresponding dynamic environment monitoring product / system includes in hardware: a dynamic environment host with edge computing capabilities, an air conditioning linkage device connected to the dynamic environment host, a multi-in-one gas sensor with multi-component gas analysis function, a partial discharge sensor, a camera, a fan linkage device, a local control panel, a micro-environment sensor with temperature and humidity monitoring function, an access control device, etc.
[0119] Embodiment 10:
[0120] This embodiment is further refined on the basis of Embodiment 9:
[0121] The oxygen concentration sensor is an electrochemical gas sensor, and the number of the oxygen concentration sensors is greater than or equal to 2;
[0122] The sulfur hexafluoride concentration sensor is an infrared gas sensor including an air chamber, and the air chamber is provided with an air inlet and an air outlet, and the air inlet and the air outlet can be switched with each other. In the above scheme, the air chamber is a closed place for the infrared gas sensor to detect the sulfur hexafluoride concentration, and the air inlet and the air outlet can be switched with each other, which is a configuration of limiting the air chamber. With this scheme, the air inlet and the air outlet can be switched to change the air inlet position of the air chamber, so that the infrared gas sensor can monitor the sulfur hexafluoride gas concentration collected at different directions around it at its installation position.
[0123] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific embodiments of the present invention are limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, other embodiments obtained without departing from the technical solution of the present invention should be included in the protection scope of the present invention.
Claims
1. A gas monitoring method for a power distribution room, the method monitoring the gas phase concentration of a target gas in the power distribution room based on a gas sensor, the gas sensor comprising at least one of an oxygen concentration sensor and a sulfur hexafluoride concentration sensor, characterized in that: In the method, the collected values of the gas sensor are collected and the environmental parameters at the time of collecting the collected values are obtained, wherein the environmental parameters include the temperature parameters and the humidity parameters in the power distribution room; The collected values and environmental parameters are input into a gas phase concentration prediction model. The gas phase concentration prediction model predicts the change law of the gas phase concentration of the target gas in the power distribution room within a preset time period according to the collected values and environmental parameters, and obtains a relationship curve between the gas phase concentration of the target gas and time within the preset time period. In the relationship curve, the time when the collected values are collected is used as the initial time, and the collected values are used as the gas phase concentration at the initial time. The obtained relationship curve is the relationship curve under the current environmental parameters; During the preset time period, according to the set gas sensor data collection rules, the gas phase concentration of the target gas in the power distribution room is monitored to obtain the actual concentration value of the target gas; In the relationship curve, the corresponding predicted gas phase concentration value is searched according to the actual concentration value acquisition time, and the actual concentration value is compared with the predicted gas phase concentration value. When the comparison result deviates from the set deviation threshold, it is determined that the gas concentration in the power distribution room is abnormal; Wherein, during the preset time period, the ventilation and access control conditions of the power distribution room are monitored. When it is detected that the access control is opened or the ventilation system is started, the collected actual concentration value is compared with the preset target gas limit concentration threshold. When the comparison result is that the actual concentration value exceeds the target gas limit concentration threshold, it is determined that the gas concentration in the power distribution room is abnormal; When it is detected that the door access control is closed and the ventilation system is closed, the collected values and environmental parameters are collected again, and a new relationship curve is obtained through the gas phase concentration prediction model; The method for the gas phase concentration prediction model to obtain the relationship curve is: according to the input collection value and environmental parameters, a sub-model closest to the current environmental parameters is searched from multiple sub-models in the database, different sub-models are configured with different environmental parameters, and the sub-model has the change law of the gas phase concentration of the target gas under the configured environmental parameters; The collected value is used as the gas phase concentration at the initial moment of the relationship curve, and the gas phase concentration at other moments in the preset time period is generated according to the change law in the sub-model found; According to the gas phase concentration at the initial moment and the gas phase concentration at other moments, the relationship curve is obtained by curve fitting; The sub-model is updated during the use of the power distribution room, and the updating method is: within the preset time period, when the comparison result between the actual concentration value of the target gas in the power distribution room at each moment and the corresponding predicted gas phase concentration value in the relationship curve is within the set deviation threshold, and the door opening and ventilation system startup are not detected within the preset time period, the change law in the sub-model found is updated according to the change law of the actual concentration value within the preset time period; The deviation threshold includes a first deviation threshold and a second deviation threshold, the first deviation threshold is used to compare the actual oxygen concentration value with the predicted oxygen concentration value, and the second deviation threshold is used to compare the actual sulfur hexafluoride concentration value with the predicted sulfur hexafluoride concentration value; For the actual oxygen concentration value collected, within a preset time period, when the actual concentration value currently collected is greater than the actual concentration value collected last time, the collected value and the environmental parameters are collected again, and a new relationship curve is obtained through the gas phase concentration prediction model; when the actual concentration value currently collected is less than the actual concentration value collected last time, the actual concentration value is compared with the predicted gas phase concentration value using the first deviation threshold; For the actual concentration of sulfur hexafluoride collected, within a preset time period, when the current actual concentration value collected is less than the actual concentration value collected last time, the collected value and environmental parameters are recollected, and a new relationship curve is obtained through the gas phase concentration prediction model; when the current actual concentration value collected is greater than the actual concentration value collected last time, the second deviation threshold is used to compare the actual concentration value with the predicted gas phase concentration value.
2. A method for monitoring gas in a power distribution room according to claim 1, characterized in that: The sub-model is obtained through the historical usage data of the current power distribution room: the historical environmental parameters of the power distribution room are divided into multiple parameter intervals according to the value size, and the historical usage data monitored during the continuous use cycle without ventilation and access control is searched and the gas phase concentration change law of each parameter interval is assigned; The method of assigning values is as follows: for any parameter interval, multiple historical target gas gas phase concentrations are collected from historical usage data, and the environmental parameters at the time of collecting each historical target gas gas phase concentration are within the parameter interval, and the change law of the target gas gas phase concentration over time in the parameter interval is calculated by using the historical target gas gas phase concentrations; After the power distribution room is overhauled, the sub-model is reconfigured; The parameter interval is a temperature parameter interval divided according to the temperature parameter, and different parameter intervals have the same humidity parameter range. In the process of searching the sub-model, the sub-model is searched according to the temperature parameter. When the humidity parameter input to the gas phase concentration prediction model belongs to the outside of the humidity parameter range, the gas phase concentration prediction model triggers the power distribution room to adjust the humidity; Different sub-models are configured with different temperature parameter ranges.
3. A method for monitoring gas in a power distribution room according to claim 1, characterized in that: The oxygen concentration sensor is an electrochemical gas sensor; The number of the oxygen concentration sensors is greater than or equal to 2, and different oxygen concentration sensors are arranged at different positions in the distribution room. The gas phase concentration prediction model is configured with a relationship curve for each oxygen concentration sensor. When the actual concentration value collected by any oxygen concentration sensor is compared to an abnormal gas concentration in the distribution room, it is determined that the gas concentration in the distribution room is abnormal.
4. A method for monitoring gas in a power distribution room according to claim 1, characterized in that: The sulfur hexafluoride concentration sensor is an infrared gas sensor, and the method for obtaining the collected value by the sulfur hexafluoride concentration sensor is: S1. Determine the location of the leakage point where sulfur hexafluoride may leak in the power distribution room; S2. According to the distribution of the leakage points in the power distribution room, the sulfur hexafluoride deposition area located below the leakage point is marked in the power distribution room; S3. The sulfur hexafluoride concentration sensor obtains the actual concentration value of sulfur hexafluoride from the sulfur hexafluoride deposition area according to the set data collection rules.
5. A method for monitoring gas in a power distribution room according to claim 4, characterized in that: When the number of the sulfur hexafluoride deposition areas is greater than or equal to 2, each sulfur hexafluoride deposition area is equipped with a sulfur hexafluoride concentration sensor; The gas phase concentration prediction model is configured with a relationship curve for each sulfur hexafluoride deposition area. When the actual concentration value collected from any sulfur hexafluoride precipitation area is compared to the abnormal gas concentration in the distribution room, it is determined that the gas concentration in the distribution room is abnormal.
6. A gas monitoring system for a power distribution room, characterized in that: The system is used to implement the method described in any one of claims 1 to 5; The system includes a gas sensor and a temperature and humidity sensor, wherein the gas sensor includes at least one of an oxygen concentration sensor and a sulfur hexafluoride concentration sensor; The system includes a data processing module, in which a gas phase concentration prediction model is provided, and the gas sensor and the temperature and humidity sensor are both data-connected to the data processing module. The gas phase concentration prediction model predicts the change law of the gas phase concentration of the target gas in the power distribution room within a preset time period according to the collected values of the gas sensor and the environmental parameters collected by the temperature and humidity sensor, and obtains a relationship curve between the gas phase concentration of the target gas and time within the preset time period, in which the collection time of the collected value is used as the initial time, and the collected value is used as the gas phase concentration at the initial time; During the preset time period, the gas sensor monitors the gas phase concentration of the target gas in the power distribution room according to the set data collection rules to obtain the actual concentration value of the target gas; In the relationship curve, the gas phase concentration prediction model searches for the corresponding predicted gas phase concentration value according to the actual concentration value acquisition time, and compares the actual concentration value with the predicted gas phase concentration value. When the comparison result deviates from the set deviation threshold, it is determined that the gas concentration in the power distribution room is abnormal; Among them, in the preset time period, the data processing module monitors the ventilation and access control conditions of the power distribution room. When it is detected that the access control is opened or the ventilation system is started, the collected actual concentration value is compared with the preset target gas limit concentration threshold. When the comparison result is that the actual concentration value exceeds the target gas limit concentration threshold, it is determined that the gas concentration in the power distribution room is abnormal; When the data processing module detects that the door access is closed and the ventilation system is closed, the collected values and environmental parameters are recollected, and a new relationship curve is obtained through the gas phase concentration prediction model.
7. The gas monitoring system for power distribution room according to claim 6, characterized in that: The oxygen concentration sensor is an electrochemical gas sensor, and the number of the oxygen concentration sensors is greater than or equal to 2; The sulfur hexafluoride concentration sensor is an infrared gas sensor comprising an air chamber, and the air chamber is provided with an air inlet and an air outlet, and the air inlet and the air outlet are switchable with each other.
8. The gas monitoring system for power distribution room according to claim 6, characterized in that: The data processing module is arranged on the dynamic environment host for monitoring the distribution room, the gas sensor is integrated on the all-in-one gas sensor, the temperature and humidity sensor is integrated on the microenvironment sensor, and also includes an intelligent door lock as a component of the access control of the distribution room, and a fan linkage device for realizing automatic operation of the ventilation system of the distribution room. The all-in-one gas sensor, microenvironment sensor, intelligent door lock and fan linkage device all communicate with the dynamic environment host so that the dynamic environment host can obtain the collected values, environmental parameters, access control status and ventilation conditions.
Citation Information
Patent Citations
Intelligent power distribution room monitoring system
CN210608694U
Gas concentration predicting device and gas concentration real-time detecting and predicting system
CN106770532A
Artificial intelligent measuring method for formaldehyde concentration
CN109781809A
Method for predicting dynamic changes of multiple environmental factors in pig house
CN111368975A
Environment monitoring method and equipment for live poultry feeding of intelligent cage and medium
CN113240152A
