Power distribution room remote monitoring and fire-fighting inspection system based on cloud platform

Abstract

The invention discloses a power distribution room remote monitoring and fire-fighting inspection system based on a cloud platform, and relates to the technical field of fire-fighting safety monitoring. The system comprises a remote monitoring module, a fire safety monitoring module and an inspection task scheduling module. And the remote monitoring module acquires and stores the running state data of the power distribution equipment in real time to realize continuous monitoring. The fire safety monitoring module obtains environment data of a power distribution room through an environment sensing unit, establishes a fire prediction model through a random forest algorithm in combination with an equipment operation state, and generates a fire prediction index. And the early warning unit simulates a fire spreading process based on the fire prediction index and predicts a fire spreading direction and an influence area. And the inspection task scheduling module intelligently distributes inspection tasks and adjusts task execution strategies in real time according to the fire risk assessment result in combination with the geographic position and the working load of the inspection personnel. The fire risk is accurately predicted and the inspection task is dynamically scheduled, so that the fire-fighting potential safety hazard of the power distribution room can be timely found and eliminated.

Description

Technical Field

[0001] The present invention relates to the technical field of fire safety monitoring, and in particular to a cloud platform-based remote monitoring and fire inspection system for power distribution rooms. Background Art

[0002] With the continuous development of social economy and the acceleration of urbanization, the power distribution system, as an important part of urban infrastructure, bears the key responsibility of providing a stable and safe power supply to the society. Especially for key equipment operation and maintenance sites such as distribution rooms, it is particularly important to ensure their normal operation and timely handle potential safety hazards. Traditional distribution room operation and maintenance relies on manual inspections and regular inspections. Although these measures can ensure equipment safety to a certain extent, with the continuous expansion of the scale of the distribution system, the efficiency and accuracy of manual inspections can no longer meet the requirements of modern smart grids for rapid response, accurate early warning and automated management.

[0003] At the same time, fire accidents are one of the common emergencies that occur in power distribution equipment, which seriously threatens the safety of power distribution equipment and the surrounding environment. Fire spreads quickly and causes great losses. If early warning and measures are not taken in time, the consequences will be disastrous. Therefore, how to effectively remotely monitor the power distribution room, fire warning and intelligently dispatch inspection tasks has become an urgent need to improve the level of power safety management.

[0004] At present, most of the monitoring and inspection of power distribution rooms still rely on manual operation and traditional detection equipment, which has many defects. The traditional manual inspection method is inefficient and easily affected by human negligence and operational errors, making it difficult to detect equipment failures or potential fire risks in a timely manner. In addition, it is difficult for manual inspections to monitor the operating status of power distribution equipment in real time, resulting in insufficient early warning capabilities for emergencies such as fires, and it is difficult to take effective preventive measures at the early stage of a fire. Most of the existing fire monitoring methods rely on single environmental monitoring data or only make predictions based on historical equipment data, lacking dynamic analysis of multiple factors. The fire spread prediction model in the prior art is usually relatively simple, and it is difficult to adjust the prediction results in real time according to the changing environment and equipment status. Therefore, the prediction of the direction of fire spread and the affected area is not accurate enough, and there are great safety hazards. Traditional inspection task scheduling is usually allocated according to a fixed schedule, ignoring the actual risk level of the distribution equipment and the workload of the inspection personnel, resulting in uneven distribution of inspection tasks, and some high-risk areas may not be inspected and handled in time.

[0005] Therefore, the existing methods have great deficiencies in intelligence, accuracy, real-time and automation. It is urgent to improve the efficiency and safety of distribution room management by introducing a cloud-based intelligent monitoring and fire inspection system combined with modern data analysis, machine learning algorithms and task scheduling technology. Summary of the invention

[0006] In view of the deficiencies in the prior art, the present invention provides a remote monitoring and fire inspection system for distribution rooms based on a cloud platform, which solves the problems of the above-mentioned background technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: a remote monitoring and fire inspection system for a power distribution room based on a cloud platform, comprising the following modules: a remote monitoring module, a fire safety monitoring module, and an inspection task scheduling module; the remote monitoring module is used to obtain the operating status data of the power distribution equipment in real time, and store the operating status data of the power distribution equipment through the cloud platform; the fire safety monitoring module includes an environmental sensing unit and an early warning unit; the environmental sensing unit is used to obtain the environmental data of the power distribution room, and in combination with the real-time operating status data and historical operating status data of the power distribution equipment, a fire prediction model is established through a random forest algorithm, the correlation pattern between the environmental data of the power distribution room and the operating status of the power distribution equipment is analyzed, and a fire prediction index is obtained; The early warning unit is used to simulate the fire spread process and dynamically predict the fire expansion direction and affected area according to the distribution room environmental data when the fire prediction index is greater than the set threshold; the inspection task scheduling module includes a task generation unit, a task scheduling unit and a task tracking unit; the task generation unit is used to evaluate the fire risk level of each distribution room according to the fire spread prediction result and the fire prediction index, and generate inspection tasks according to the priority of the fire risk assessment; the task scheduling unit is used to allocate inspection tasks according to the inspection task priority, combined with the geographical location and workload of the inspection personnel; the task tracking unit is used to record the movement trajectory of the inspection personnel, and adjust the execution strategy of the inspection task according to the feedback information of the inspection personnel.

[0008] Furthermore, the operating status data of the power distribution equipment includes equipment temperature data, current and voltage data, and equipment load data; the environmental data of the power distribution room includes temperature and humidity data, smoke concentration data, and combustible gas concentration data.

[0009] Furthermore, the specific process of establishing a fire prediction model through the random forest algorithm is as follows: preprocessing is performed based on the operating status data of the distribution equipment and the environmental data of the distribution room, including noise removal, missing value filling, and normalization processing, to extract features related to fire risk, including temperature fluctuations of the distribution equipment, changes in environmental humidity, and current anomalies, and to construct a fire risk feature set; the processed data is divided into a training set and a test set according to the time series, the training set is used for model construction, and the test set is used to verify the model effect; based on the training set data, multiple decision tree models are constructed through the random forest algorithm. Each decision tree establishes a fire risk prediction model by selecting and dividing different features.

[0010] Furthermore, the correlation pattern between the distribution room environmental data and the operating status of the distribution equipment is analyzed. The specific process of obtaining the fire prediction index is as follows: according to the fire risk feature set, fire risk-related features are extracted from the training set and the test set, and data preprocessing is performed; the relationship between the distribution room environmental data and the distribution equipment operating status data is analyzed by the Pearson correlation coefficient, and the characteristics that affect the fire risk are identified and weighted combination is performed; the weighted feature data is used to calculate the fire risk index and obtain the fire prediction index.

[0011] Furthermore, based on the environmental data of the distribution room, the fire spread process is simulated, and the specific process of dynamically predicting the expansion direction and affected area of ​​the fire is as follows: Based on the environmental data of the distribution room, a fire spread model is established, and the physical model and algorithm of the fire spread are defined in combination with the building structure and environmental factors of the distribution room; by simulating the fire spread process, the fire expansion speed, spread direction and affected area are calculated, and the fire spread trend map is obtained; the fire spread model is dynamically adjusted according to the real-time environmental data and fire prediction index, and the future development trend of the fire is predicted; the area and risk level of the fire spread are determined based on the prediction results.

[0012] Furthermore, based on the fire spread prediction results and the fire prediction index, the specific process of evaluating the fire risk level of each distribution room is as follows: the fire spread trend chart and the fire prediction index are comprehensively processed to obtain the fire risk index of each distribution room, which is used to evaluate the fire risk level of each distribution room.

[0013] Furthermore, according to the priority of fire risk assessment, the specific process of generating inspection tasks is as follows: according to the fire risk index of each distribution room, the fire risk priority of each distribution room is determined in order from high risk to low risk; based on the risk priority, an inspection task is generated for each distribution room to determine the equipment, time, and task content information that need to be inspected.

[0014] Furthermore, according to the inspection task priority, combined with the geographical location and workload of the inspection personnel, the specific process of allocating inspection tasks is as follows: obtain the current location, workload and available time of the inspection personnel, calculate the geographical distance between each inspection personnel and each distribution room according to the inspection task priority, and evaluate whether the task can be completed in combination with the workload and available time of the inspection personnel; assign the inspection task to the inspection personnel who is closest and can complete the inspection task.

[0015] The present invention has the following beneficial effects: (1) The remote monitoring and fire inspection system for power distribution rooms based on the cloud platform effectively improves the operating status monitoring and fire warning capabilities of power distribution equipment through the collaborative work of the remote monitoring module and the fire safety monitoring module. The remote monitoring module can obtain the operating status data of the power distribution equipment in real time, and store and analyze it through the cloud platform to achieve comprehensive monitoring of the power distribution equipment and detect potential equipment failures and abnormal conditions in advance. The fire safety monitoring module combines the environmental sensing unit and the early warning unit, and uses the random forest algorithm to analyze the environmental data and equipment status of the power distribution room to generate an accurate fire prediction index. When the fire prediction index exceeds the set threshold, the system can promptly start the fire spread simulation to provide data support for early warning and prevention of fires.

[0016] (2) The cloud-based remote monitoring and fire inspection system for distribution rooms has significantly improved the management efficiency and accuracy of inspection tasks through the inspection task scheduling module. The task generation unit evaluates the fire risk of each distribution room based on the fire spread prediction results and fire prediction index, and generates inspection tasks with clear priorities based on the risk level. The task scheduling unit intelligently allocates tasks based on the priority of the inspection task, the geographical location and workload of the inspection personnel, ensuring that inspection resources are reasonably allocated and improving inspection efficiency. The task tracking unit records the movement trajectory of the inspection personnel and dynamically adjusts the inspection strategy based on the feedback information to ensure the smooth execution of the inspection task, further improving the safety management level of the distribution room.

[0017] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a flow chart of the remote monitoring and fire inspection system for power distribution rooms based on a cloud platform of the present invention. DETAILED DESCRIPTION

[0019] The embodiment of the present application solves the problems of insufficient data integration, delayed fire prediction, and unreasonable scheduling of inspection tasks in existing distribution equipment monitoring and fire inspections through a remote monitoring and fire inspection system for distribution rooms based on a cloud platform. By combining the operating status data of distribution equipment with environmental data and applying the random forest algorithm to establish a fire prediction model, the system can assess fire risks in real time and predict the trend of fire spread. At the same time, combined with the priority of inspection tasks, the geographical location and workload of inspection personnel, inspection tasks are intelligently scheduled, which improves inspection efficiency and fire response capabilities, thereby effectively improving the safety management level of distribution rooms.

[0020] The overall idea of ​​the solution in the embodiments of this application is as follows: Obtain the operating status data of power distribution equipment in real time and store it through the cloud platform The environmental data of the distribution room is obtained, and combined with the real-time operating status data and historical operating status data of the distribution equipment, a fire prediction model is established through the random forest algorithm, the correlation pattern between the environmental data of the distribution room and the operating status of the distribution equipment is analyzed, and the fire prediction index is obtained.

[0021] When the fire prediction index is greater than the set threshold, the fire spread process is simulated based on the distribution room environmental data, and the fire expansion direction and affected area are dynamically predicted.

[0022] According to the fire spread prediction results and fire prediction index, the fire risk level of each distribution room is evaluated, and inspection tasks are generated according to the priority of fire risk assessment.

[0023] Inspection tasks are assigned based on inspection task priority, combined with the inspectors’ geographic location and workload.

[0024] Record the movement trajectory of patrol personnel and adjust the execution strategy of patrol tasks based on their feedback.

[0025] See also Figure 1 The embodiment of the present invention provides a technical solution: a remote monitoring and fire inspection system for a power distribution room based on a cloud platform, comprising the following modules: a remote monitoring module, a fire safety monitoring module, and an inspection task scheduling module; the remote monitoring module is used to obtain the operating status data of the power distribution equipment in real time, and store the operating status data of the power distribution equipment through the cloud platform; the fire safety monitoring module includes an environmental sensing unit and an early warning unit; the environmental sensing unit is used to obtain the environmental data of the power distribution room, and in combination with the real-time operating status data and historical operating status data of the power distribution equipment, a fire prediction model is established through a random forest algorithm, the correlation pattern between the environmental data of the power distribution room and the operating status of the power distribution equipment is analyzed, and a fire prediction index is obtained; the early warning The unit is used to simulate the fire spread process and dynamically predict the fire expansion direction and affected area according to the distribution room environmental data when the fire prediction index is greater than the set threshold; the inspection task scheduling module includes a task generation unit, a task scheduling unit and a task tracking unit; the task generation unit is used to evaluate the fire risk level of each distribution room according to the fire spread prediction results and the fire prediction index, and generate inspection tasks according to the priority of the fire risk assessment; the task scheduling unit is used to allocate inspection tasks according to the inspection task priority, combined with the geographical location and workload of the inspection personnel; the task tracking unit is used to record the movement trajectory of the inspection personnel, and adjust the execution strategy of the inspection tasks according to the feedback information of the inspection personnel.

[0026] In this implementation scheme, remote monitoring module: This module is responsible for obtaining the operating status data of the power distribution equipment in real time. The real-time status of the power distribution equipment is collected by sensors and stored and managed through the cloud platform. Workflow: Obtain the real-time operating data of the power distribution equipment. Upload the data to the cloud platform for storage and subsequent analysis. The power distribution equipment can be remotely monitored through the cloud platform to ensure that the operating status of the equipment meets safety standards. Fire safety monitoring module, which includes an environmental sensing unit and an early warning unit, is mainly used to monitor the environmental data of the distribution room in real time and predict the fire risk through algorithms. Environmental sensing unit: Function: Responsible for obtaining environmental data in the distribution room, such as temperature, humidity, gas concentration, etc. In addition, it also collects the real-time operating status and historical operating data of the distribution equipment. By combining these data with the associated features of fire occurrence (such as equipment temperature, humidity changes, etc.), a fire prediction model is established using machine learning algorithms (such as random forest algorithms). Workflow: Collect real-time data of the environment and equipment. Use the random forest algorithm to process these data and extract fire risk-related features. Calculate the fire prediction index based on the trained model to provide a preliminary assessment of the fire risk. Early warning unit: Function: According to the data provided by the environmental sensing unit, when the fire prediction index exceeds the set threshold, the early warning mechanism is activated. And further simulate the fire spread process to predict the direction of fire expansion and the affected area. Workflow: Monitor the fire prediction index, and trigger the early warning if it exceeds the threshold. Based on environmental data, simulate the dynamic process of fire spread. Predict the expansion speed, spread direction and impact range of the fire. Inspection task scheduling module This module is responsible for automatically generating inspection tasks based on the fire risk assessment results and reasonably scheduling the work of inspection personnel. Task generation unit: Function: According to the fire spread prediction results and fire prediction index of the distribution room, evaluate the fire risk level of each distribution room, and generate inspection tasks according to the risk priority. High-risk distribution rooms will be given priority for inspection. Workflow: Combine the fire prediction index with the spread prediction results to calculate the fire risk index of each distribution room. Determine the priority of the inspection task according to the fire risk index. Generate inspection tasks according to the priority, including inspection time, task content and inspection equipment. Task scheduling unit: Function: According to the priority of the inspection task, combined with the geographical location, workload and available time of the inspection personnel, the inspection tasks are reasonably allocated. Workflow: Obtain the geographical location information and workload (number of completed tasks, remaining tasks, etc.) of the inspection personnel. Assign tasks according to the priority of the inspection task and the free time of the inspection personnel. Ensure that the inspection tasks in high-risk areas can be assigned to the most suitable inspection personnel first. Task tracking unit: Function: Record the movement trajectory of the inspection personnel, track the execution progress of the task in real time, and adjust the task according to the feedback information to ensure the effective completion of the task. Workflow: Monitor the task progress and work status of the inspection personnel in real time.Collect feedback from inspectors and adjust task content or inspectors' work arrangements to respond to emergencies or task changes.

[0027] Specifically, the operating status data of the power distribution equipment includes equipment temperature data, current and voltage data, and equipment load data; the distribution room environment data includes temperature and humidity data, smoke concentration data, and combustible gas concentration data.

[0028] In this implementation scheme, the distribution equipment operation status data: Equipment temperature data: Monitoring the temperature of the distribution equipment is one of the key indicators for determining whether the equipment is overloaded. Excessive temperature may indicate that the equipment is at risk of overheating, resulting in malfunction or fire hazard. Current and voltage data: Current and voltage data are used to monitor the power operation of the equipment in real time. Abnormal current or voltage fluctuations may cause equipment damage or even fire. Equipment load data: The load condition of the distribution equipment is an important parameter for evaluating whether the equipment is working properly and whether there is an overload risk. Excessive load may cause equipment damage or fire accidents. Distribution room environment data: Temperature and humidity data: Temperature and humidity data help understand the environmental status of the distribution room. Abnormal temperature and humidity may have an adverse effect on the equipment, thereby increasing the risk of fire. For example, high temperature and high humidity environments are prone to cause electrical equipment failures. Smoke concentration data: Smoke concentration detection can promptly detect smoke in the early stages of a fire, thereby providing early warning. Combustible gas concentration data: Excessive concentration of combustible gas may indicate that there are dangers such as leakage in the distribution room. Timely detection of combustible gas leakage helps prevent fires.

[0029] Specifically, the specific process of establishing a fire prediction model through the random forest algorithm is as follows: pre-processing is performed based on the operating status data of the distribution equipment and the environmental data of the distribution room, including noise removal, missing value filling, and normalization processing, to extract features related to fire risk, including temperature fluctuations of the distribution equipment, changes in environmental humidity, and current anomalies, and to construct a fire risk feature set; the processed data is divided into a training set and a test set according to the time series, the training set is used for model construction, and the test set is used to verify the model effect; based on the training set data, multiple decision tree models are constructed through the random forest algorithm. Each decision tree establishes a fire risk prediction model by selecting and dividing different features.

[0030] In this implementation scheme, data preprocessing: Noise removal: Remove outliers and noise points in the data to avoid these abnormal data affecting the training effect of the model. Common sources of noise include sensor errors, data entry errors, etc. Fill missing values: For missing data, appropriate methods (such as mean filling, interpolation, etc.) are usually used to fill them. If missing values ​​are not processed, it will lead to incomplete data during model training and affect the accuracy of prediction. Normalization processing: Convert data of different dimensions and scales (such as temperature, current, humidity, etc.) into a unified standard range (such as 0 to 1) to avoid certain features from having too much impact on model training and ensure the training stability of the model. Feature extraction: Extract features related to fire risk, usually based on the operating status of the distribution equipment and the environmental data of the distribution room to judge the fire risk. Specific features include: Temperature fluctuations of distribution equipment: Equipment with large temperature changes may have overheating or failure risks. Changes in environmental humidity: Excessive humidity may affect the performance of electrical equipment, leading to fire hazards such as short circuits. Abnormal current: Sudden changes in current or exceeding the normal range may be a signal of equipment failure or short circuit. These feature data are used to construct a fire risk feature set to provide key input for subsequent modeling. Divide the training set and test set: According to the time series of the data, the processed data is divided into a training set and a test set. The training set is used to build the model, and the test set is used to test the prediction ability of the model. The division of the time series can ensure the timeliness of the model and the independence of the data, and avoid the leakage of future data into the training process. Use the random forest algorithm to build a decision tree model: The random forest algorithm performs ensemble learning by building multiple decision trees. Each decision tree randomly selects features during the training process and makes decisions by selecting and dividing these features. At each tree node, the algorithm selects an optimal feature (such as temperature, humidity, current, etc.) for division until the specified tree depth is reached or a certain stopping criterion is reached. This method generates the final prediction result through the voting of multiple independent decision trees, which improves the generalization ability and accuracy of the model.

[0031] Specifically, the correlation pattern between the distribution room environmental data and the operating status of the distribution equipment is analyzed, and the specific process of obtaining the fire prediction index is as follows: according to the fire risk feature set, fire risk-related features are extracted from the training set and the test set, and data preprocessing is performed; the relationship between the distribution room environmental data and the distribution equipment operating status data is analyzed by the Pearson correlation coefficient, and the characteristics that affect the fire risk are identified and weighted combinations are performed; the weighted feature data is used to calculate the fire risk index and obtain the fire prediction index.

[0032] In this implementation scheme, in the process of analyzing the correlation pattern between the distribution room environmental data and the operating status of the distribution equipment and obtaining the fire prediction index, the key steps and formulas are as follows: Data preprocessing and feature extraction: Fire risk feature set: According to the operating status of the distribution equipment (such as temperature, current, voltage, load, etc.) and the environmental data of the distribution room (such as temperature and humidity, smoke concentration, combustible gas concentration, etc.), extract the characteristic data related to the fire risk. These characteristic data contain indicators related to the possibility of fire and organize them into feature sets. · Data preprocessing: Perform conventional data preprocessing steps such as noise removal, missing value filling, and normalization to ensure the quality and availability of the data. Pearson correlation coefficient analysis: The Pearson correlation coefficient is used to measure the linear correlation between two variables. In this method, the Pearson correlation coefficient is used to analyze the relationship between the distribution room environmental data and the distribution equipment operating status data. Through this analysis, it is possible to identify which features have a strong impact on fire risk prediction and provide a basis for subsequent weighted combination Pearson correlation coefficient formula; ;in: Environmental data for the power distribution room and distribution equipment operating status data The Pearson correlation coefficient between They are the environmental data of the power distribution room and the operating status data of the power distribution equipment. data points, It is the average of the distribution room environment data and the distribution equipment operation status data. is the number of samples in the data set. The value range of the Pearson correlation coefficient is [-1,1], where Indicates a perfect positive correlation; indicates a completely negative correlation. Indicates irrelevant. Weighted feature combination: Based on the calculation results of the Pearson correlation coefficient, select the features that are most closely related to fire risk and assign weights to these features. The weight of the features can be weighted according to their relevance to fire risk. The features with stronger relevance are given higher weights. Features The weight of For the The fire prediction index is calculated by the weighted fire risk feature set. The fire prediction index can be expressed as the following formula: ;in: It is the fire prediction index, which is used to indicate the possibility of fire. Fire risk characteristics The weight of Fire risk characteristics The value of is the number of features, is the data of the environmental factors related to fire prediction (temperature and humidity, smoke concentration, gas concentration), q represents the total number of environmental factor data, and are weight coefficients, which are used to adjust the fire risk index and the impact of environmental factors on fire prediction. The higher the value, the greater the probability of fire. Set a threshold in the model like , then trigger the fire warning; if , it means that the risk of fire is low and routine monitoring can be carried out.

[0033] Specifically, based on the environmental data of the distribution room, the fire spread process is simulated and the specific process of dynamically predicting the expansion direction and affected area of ​​the fire is as follows: Based on the environmental data of the distribution room, a fire spread model is established, and the physical model and algorithm of the fire spread are defined in combination with the building structure and environmental factors of the distribution room; by simulating the fire spread process, the fire expansion speed, spread direction and affected area are calculated, and the fire spread trend map is obtained; the fire spread model is dynamically adjusted according to the real-time environmental data and fire prediction index, and the future development trend of the fire is predicted; the area and risk level of the fire spread are determined based on the prediction results.

[0034] In this implementation scheme, the fire spread process is simulated according to the environmental data of the distribution room, and the specific process of dynamically predicting the expansion direction of the fire and the affected area is as follows, and a fire spread model is established: Environmental data and building structure analysis: The fire spread process in the distribution room is affected by multiple factors, including the environmental conditions inside the distribution room (such as temperature, humidity, smoke concentration, airflow, etc.) and the building structure (wall material, windows, door layout, ventilation facilities, etc.). These environmental factors and building structures will affect the speed and direction of fire spread. Therefore, it is first necessary to build a physical model of fire spread based on these data. Physical model and algorithm: The physical model of fire spread is usually modeled using physical principles such as heat conduction, convection and radiation. Common physical models include heat conduction equations and fluid dynamics models. For mathematical modeling of fire spread, a two-dimensional heat conduction model or a CFD (computational fluid dynamics) model can be used to describe the temperature changes of the fire and its propagation process. For example, the two-dimensional heat conduction equation can be expressed as: ;in, is the temperature field, It's time. is the thermal diffusivity, It is a heat source term. Spread algorithm design: The fire spread algorithm can use a rule-based model (for example, the fire spreads from one location to an adjacent location within each time step), or use a machine learning algorithm (such as a regression model based on training data) to predict the speed and direction of fire spread. Simulate the fire spread process: Calculate the fire spread speed and spread direction: After the fire spread model is established, the fire spread process is simulated to calculate the fire spread speed and spread direction at each time point. The fire spread speed refers to the temperature and heat propagation speed in the fire source area. The spread direction is affected by the structure of the distribution room, airflow, and the characteristics of the fire source itself. The spread speed can be estimated by the energy release of the fire source and the heat conduction of the surrounding environment. Fire spread speed It can be expressed by the following formula ;in, is the intensity of the heat source, is the air density, is the specific heat capacity, is the thermal conductivity. Calculation of the affected area: The affected area of ​​fire spread is the spatial range affected by the spread of fire. As the fire spreads, heat and smoke will spread to the surrounding area, which may cause failures of other equipment or produce toxic gases. Therefore, it is necessary to calculate the affected area of ​​the fire in combination with the fire spread speed, direction and fire source intensity. The fire spread area can be obtained through mathematical simulation and can be displayed in the form of heat map, etc., showing the change of the fire spread range over time. Obtaining the fire spread trend map: Fire spread trend map: By simulating the fire spread process, the fire spread trend map can be obtained. The trend map shows the expansion of the fire in different time periods, including the speed, direction and impact range of the fire spread. Through these trend maps, managers can clearly understand the development of the fire and help make real-time decisions and scheduling. Including, the trend map can show the time and spatial distribution of the fire spread, usually in the form of heat map, isotherm map, etc., indicating the speed of the fire spread and the possible expansion area. Dynamic adjustment of the fire spread model: Real-time environmental data update: The environmental conditions of the distribution room will change over time, and factors such as temperature, humidity, and smoke concentration may change due to changes in the ventilation system or the external environment. In order to improve the accuracy of the simulation, it is necessary to dynamically adjust the fire spread model based on the real-time collected environmental data (such as temperature, humidity, airflow and other data obtained by sensors). Feedback of fire prediction index: The fire prediction index (based on environmental data, distribution equipment operation status, etc.) is also a key parameter. When the prediction index is higher than a certain set threshold, the model may need to adjust some parameters in the fire spread process, such as the spread speed, fire source intensity, etc. Real-time feedback adjustment: At each time step, the parameters of fire spread are dynamically adjusted through real-time feedback data. For example, if the smoke concentration or temperature at a certain location rises sharply, the model will adjust the spread speed and direction accordingly to predict the further expansion of the fire. Predict the future development trend of the fire: Prediction of future development trends: Based on the simulated and real-time updated fire spread model, predict the development trend of the fire in the next few days. This process needs to take into account various dynamic factors of fire spread (such as environmental changes, fire source location, building structure, etc.) and estimate the spread in the future. Determination of risk areas and levels: Based on the prediction results, the affected areas of fire spread are delineated and their risk levels are evaluated. The fire risk level can be calculated based on a weighted calculation of multiple factors (such as temperature, smoke concentration, gas concentration, etc.) and can be expressed by the fire prediction formula: ;in: For fire spreading areas at time the risk level; is the weight coefficient, which indicates the influence of different environmental factors on risk; is the smoke concentration; is the combustible gas concentration; is the fire area temperature; The fire spread area.

[0035] Specifically, according to the fire spread prediction results and the fire prediction index, the specific process of evaluating the fire risk level of each distribution room is as follows: the fire spread trend chart and the fire prediction index are comprehensively processed to obtain the fire risk index of each distribution room, which is used to evaluate the fire risk level of each distribution room.

[0036] In this implementation plan, the fire risk level of each distribution room is evaluated by comprehensively analyzing the fire spread prediction results and the fire prediction index. The specific process is as follows: Combining the fire spread trend chart and the fire prediction index: First, combine the fire spread trend chart (that is, the speed of fire spread, the direction of spread and the affected area) with the fire prediction index (fire risk based on environmental data and equipment status assessment). The fire spread trend chart provides spatial and temporal information on the dynamic expansion of the fire, while the fire prediction index reflects the possibility and severity of the fire. Comprehensive calculation: Perform a comprehensive calculation on the fire spread trend chart and the fire prediction index to obtain a numerical value that reflects the comprehensive fire risk, namely the fire risk index. Fire risk index The calculation formula is: ;in: Represents the fire risk index of the i-th distribution room. It represents the fire prediction index of the ith distribution room calculated based on environmental data and equipment operation status. The index combines the factors of equipment temperature, humidity, and current abnormality to reflect the probability of fire. The fire expansion speed of the i-th distribution room reflects the rate and expansion range of the fire spread. is the risk level value of the fire spread area in the i-th distribution room. : These are weight coefficients used to adjust the relative importance of each parameter in the fire risk index. The selection of weight coefficients is based on historical data analysis and actual conditions and is determined through the analytic hierarchy process.

[0037] Specifically, according to the priority of fire risk assessment, the specific process of generating inspection tasks is as follows: according to the fire risk index of each distribution room, determine the fire risk priority of each distribution room in order from high risk to low risk; based on the risk priority, generate inspection tasks for each distribution room, and determine the equipment that needs to be inspected, time, and task content information.

[0038] In this implementation plan, the risk priority is determined based on the fire risk index: First, the distribution rooms are sorted from high to low based on the fire risk index of each distribution room. Distribution rooms with higher fire risk indexes have a greater risk of fire and should be handled as a priority. Generate inspection tasks: Based on the determined fire risk priority, the system generates inspection tasks for each distribution room. Inspection tasks include the following: Equipment that needs to be inspected: Based on the results of the fire risk assessment, determine the equipment that needs to be inspected, such as the temperature and current of the distribution equipment. Time: According to the urgency of the fire risk, arrange a suitable time for the inspection. Task content information: Specific inspection items, including equipment status inspection, inspection of fire protection facilities, etc. This step ensures that inspection tasks can be started in a timely manner and effectively executed in distribution rooms with higher fire risks.

[0039] Specifically, according to the inspection task priority, combined with the inspection personnel's geographical location and workload, the specific process of allocating inspection tasks is as follows: obtain the inspection personnel's current location, workload and available time, calculate the geographical distance between each inspection personnel and each distribution room according to the inspection task priority, and evaluate whether the task can be completed based on the inspection personnel's workload and available time; assign the inspection task to the inspection personnel who is closest and can complete the inspection task.

[0040] In this implementation plan, the basic information of the inspection personnel is obtained: Current location: The current location of the inspection personnel is obtained in order to calculate the distance to each distribution room. Workload: Understand the amount of tasks that each inspection personnel has completed to ensure that the workload is reasonably distributed. Available time: Obtain the available time of the inspection personnel to determine whether they have enough time to perform new inspection tasks. Calculate the geographical distance between the inspection personnel and each distribution room: According to the current location of the inspection personnel and the geographical location of the distribution room, calculate the actual distance between each inspection personnel and each distribution room. This step ensures that the inspection task can be preferentially assigned to the inspection personnel closest to the distribution room to reduce the response time. Evaluate workload and available time: Combine the current workload and available time of the inspection personnel to evaluate whether each inspection personnel can complete the inspection task within the specified time. For inspection personnel with a lighter workload and sufficient time, tasks are assigned preferentially. Task allocation: Assign inspection tasks to qualified inspection personnel, that is, the inspection personnel closest to the distribution room. Inspection personnel with a lighter workload and sufficient time to complete the inspection task.

[0041] In summary, this application has at least the following effects: The remote monitoring and fire inspection system for power distribution rooms based on cloud platforms can obtain the operating status and environmental data of power distribution equipment in real time through the remote monitoring and fire inspection system based on cloud platforms. It can effectively predict fire risks, take preventive measures in advance, reduce the possibility of fire, and ensure the safety of power distribution rooms by combining advanced fire prediction models and fire spread simulation technologies. By using the fire risk assessment results and the inspection task scheduling module, combined with the geographical location, workload and available time of the inspection personnel, the priority sorting and intelligent allocation of inspection tasks are realized to ensure the timeliness and effectiveness of inspection tasks, avoid excessive concentration or task delays, and improve resource utilization and inspection efficiency. Through the task tracking module, the action trajectory and feedback information of the inspection personnel are recorded in real time, and the task execution strategy is dynamically adjusted according to the actual execution situation, which improves the flexibility and adaptability of the system and helps to achieve long-term and stable inspection management. The comprehensive environmental data, equipment operation data and fire risk analysis results of the power distribution room provide comprehensive decision-making support for managers, effectively improve the intelligent level of fire management in the power distribution room, and provide reliable guarantee for safe production.

[0042] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0043] The present invention is described with reference to flowcharts and / or block diagrams of systems, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0044] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1A function specified in one or more boxes.

[0045] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0046] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0047] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. The remote monitoring and fire inspection system for power distribution rooms based on cloud platform is characterized by: Includes the following modules: remote monitoring module, fire safety monitoring module, inspection task scheduling module; The remote monitoring module is used to obtain the operating status data of the power distribution equipment in real time and store the operating status data of the power distribution equipment through the cloud platform; The fire safety monitoring module includes an environmental sensing unit and an early warning unit; The environmental sensing unit is used to obtain the environmental data of the power distribution room, and in combination with the real-time operating status data and historical operating status data of the power distribution equipment, establish a fire prediction model through a random forest algorithm, analyze the correlation pattern between the environmental data of the power distribution room and the operating status of the power distribution equipment, and obtain a fire prediction index; The early warning unit is used to simulate the fire spreading process and dynamically predict the fire expansion direction and affected area according to the distribution room environment data when the fire prediction index is greater than the set threshold; The inspection task scheduling module includes a task generation unit, a task scheduling unit and a task tracking unit; The task generation unit is used to evaluate the fire risk level of each distribution room according to the fire spread prediction result and the fire prediction index, and generate inspection tasks according to the priority of the fire risk assessment; The task scheduling unit is used to allocate inspection tasks according to the inspection task priority, combined with the geographical location and workload of the inspection personnel; The task tracking unit is used to record the action track of the inspection personnel and adjust the execution strategy of the inspection task according to the feedback information of the inspection personnel.

2. The cloud platform-based power distribution room remote monitoring and fire inspection system according to claim 1 is characterized by: The power distribution equipment operation status data includes equipment temperature data, current and voltage data, and equipment load data; The distribution room environment data includes temperature and humidity data, smoke concentration data, and combustible gas concentration data.

3. The cloud platform-based power distribution room remote monitoring and fire inspection system according to claim 2 is characterized in that: The specific process of establishing a fire prediction model through the random forest algorithm is as follows: Preprocessing is performed based on the distribution equipment operation status data and distribution room environment data, including noise removal, missing value filling, and normalization processing, to extract features related to fire risk, including temperature fluctuations, environmental humidity changes, and current anomalies of distribution equipment, and to construct a fire risk feature set; The processed data is divided into training set and test set according to the time series. The training set is used for model construction, and the test set is used to verify the model effect. Based on the training set data, multiple decision tree models are constructed using the random forest algorithm. Each decision tree establishes a fire risk prediction model by selecting and dividing different features.

4. The cloud platform-based power distribution room remote monitoring and fire inspection system according to claim 3 is characterized in that: The specific process of analyzing the correlation pattern between the distribution room environment data and the operating status of the distribution equipment and obtaining the fire prediction index is as follows: According to the fire risk feature set, fire risk related features are extracted from the training set and the test set, and data preprocessing is performed; The relationship between the distribution room environment data and the distribution equipment operation status data is analyzed through the Pearson correlation coefficient to identify the characteristics that affect the fire risk and make a weighted combination; The fire risk index is calculated using the weighted characteristic data to obtain the fire prediction index.

5. The cloud platform-based power distribution room remote monitoring and fire inspection system according to claim 4 is characterized in that: According to the distribution room environment data, the fire spread process is simulated and the specific process of dynamically predicting the fire expansion direction and affected area is as follows: According to the environmental data of the power distribution room, a fire spread model is established, and the physical model and algorithm of fire spread are defined by combining the building structure and environmental factors of the power distribution room; By simulating the fire spread process, the fire expansion speed, spread direction and affected area are calculated to obtain the fire spread trend map; Dynamically adjust the fire spread model based on real-time environmental data and fire prediction index to predict future development trends of fires; Determine the area and risk level of fire spread based on the prediction results.

6. The cloud platform-based power distribution room remote monitoring and fire inspection system according to claim 5 is characterized in that: According to the fire spread prediction results and fire prediction index, the specific process of evaluating the fire risk level of each distribution room is as follows: The fire spread trend graph and fire prediction index are comprehensively processed to obtain the fire risk index of each distribution room, which is used to evaluate the fire risk level of each distribution room.

7. The cloud platform-based power distribution room remote monitoring and fire inspection system according to claim 6 is characterized by: According to the priority of fire risk assessment, the specific process of generating inspection tasks is as follows: According to the fire risk index of each distribution room, determine the fire risk priority of each distribution room in the order of high risk to low risk; Generate inspection tasks for each distribution room based on risk priority, and determine the equipment, time, and task content information that need to be inspected.

8. The cloud platform-based power distribution room remote monitoring and fire inspection system according to claim 7 is characterized in that: According to the inspection task priority, combined with the inspection personnel's geographical location and workload, the specific process of allocating inspection tasks is as follows: Obtain the current location, workload and available time of the inspectors, calculate the geographical distance between each inspector and each distribution room according to the inspection task priority, and evaluate whether the task can be completed based on the workload and available time of the inspectors; Assign inspection tasks to the inspectors who are closest and capable of completing the inspection tasks.

Images