A substation secondary equipment intelligent storeroom system

Abstract

The application discloses a kind of intelligent storeroom systems of substation secondary equipment, including secondary equipment storeroom and secondary equipment intelligent management system;Secondary equipment storeroom is deployed sensor and camera to carry out temperature and humidity, smoke, personnel and flame detection;The RFID equipment of secondary equipment storeroom captures RFID tag data, records information such as warehousing and delivery;Face recognition scanner of secondary equipment storeroom is authenticated and access management to authorized personnel;Secondary equipment intelligent management system inquires spare parts, displays spare part information, records management to spare part warehousing and delivery, carries out inventory alarm according to spare part quantity, uses inventory data to analyze, monitors, alarms, inquires to storeroom environment data, optimizes alarm strategy, uses the optimization algorithm of catharsius to optimize adjustment to safety threshold, carries out user management to storeroom manager.The application ensures the normal operation of secondary equipment storeroom and the safety and reliability of equipment.

Description

Technical Field

[0001] This invention belongs to the field of substation engineering, and specifically relates to an intelligent warehouse system for secondary equipment in substations. Background Technology

[0002] The safe and stable operation of substation secondary equipment is directly related to the safety and stability of the entire power grid. As a crucial location for storing and housing secondary equipment, the performance of the stored equipment is directly determined. Therefore, substation secondary equipment warehouses are of great significance for ensuring the normal operation of substations and improving the reliability and security of the power grid. Substation secondary equipment warehouses typically store spare parts and replacement components to cope with sudden failures or equipment damage. The timely supply and management of these spare parts directly affects the repair and recovery time of substation equipment, and consequently, the reliability and stability of the power grid. In the event of a sudden failure or emergency, if the warehouse cannot quickly provide the necessary equipment and parts, the substation cannot be restored to operation rapidly, leading to accidents that impact the power grid and users. Traditional substation secondary equipment warehouses do not consider the storage needs of special equipment and equipment safety protection requirements, and cannot achieve equipment management and traceability, as well as intelligent spare parts management and replenishment. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides an intelligent warehouse system for secondary equipment in substations, which solves the technical problems of normal operation of secondary equipment warehouses and the safety and reliability of equipment.

[0004] To solve the above-mentioned technical problems, the present invention adopts the following technical solution.

[0005] This invention first discloses an intelligent warehouse system for substation secondary equipment, comprising a secondary equipment warehouse and a secondary equipment intelligent management system. The secondary equipment warehouse is divided into a management area, a common tools area, a testing instrument area, a spare parts area, a relay protection spare parts area, an automation spare parts area, and a network security spare parts area. The secondary equipment warehouse is equipped with sensors and cameras for temperature and humidity detection, smoke detection, personnel detection, and flame detection. The warehouse is also equipped with RFID devices to capture RFID tag data and record inbound and outbound information. Furthermore, the warehouse is equipped with facial recognition scanners for authorized personnel authentication and access management.

[0006] The secondary equipment intelligent management system is used to query spare parts, display spare parts information, record and manage the entry and exit of spare parts, issue inventory alarms based on the quantity of spare parts, analyze inventory data, monitor and query warehouse environment data, and manage warehouse management personnel.

[0007] The flame detection adopts the YOLOv8 architecture and uses the YOLO loss function. First, the YOLO model is trained on the flame detection dataset. Early stopping technique is used to prevent overfitting. The model is evaluated on the test set and mAP is used as the performance metric. Based on the performance on the test set, the optimal detection threshold is determined and the trained model is deployed to the real-time monitoring system.

[0008] Real-time flame detection is performed on images captured by the camera. The first step is image processing, including:

[0009] The video stream inside the warehouse is captured using a camera, and images within a specific wavelength range are used to enhance the characteristics of the flames. Image preprocessing includes grayscale conversion and histogram equalization.

[0010] I'(x,y)=T·I(x,y)

[0011] Where I'(x,y) is the processed image intensity, I(x,y) is the original image intensity, and T is the transformation matrix based on the flame spectral characteristics;

[0012] The processed image is then input into the trained YOLO model for flame detection.

[0013] The present invention further includes the following preferred embodiments:

[0014] Each storage area in the management area, common tools area, testing instrument area, accessories area, relay protection spare parts area, automation spare parts area, and network security spare parts area is labeled with a category name. Different areas are marked with different colors, and physical dividing lines are used to separate different categories. The management area is equipped with office computers and servers, and the intelligent management system for secondary equipment is deployed there.

[0015] The secondary equipment intelligent management system will pre-register the facial data of authorized personnel. The facial recognition scanner is integrated with the electronic door lock and uses facial recognition technology for detection. Only authorized personnel can enter the warehouse. Access permissions are managed and monitored by setting access control software to define access levels for different groups or individuals.

[0016] The intelligent management system for secondary equipment includes:

[0017] The spare parts search and query module is used to search for spare parts by name and display the spare parts information and the shelf they belong to based on the search results;

[0018] The inbound / outbound record module is used to manage inventory and view the time of equipment entry and exit.

[0019] The spare parts inventory alarm module is used to generate equipment inventory alarm records based on the equipment inventory quantity; it defines the minimum inventory level replenishment threshold for each spare part, and when the inventory is lower than the corresponding threshold, it triggers a replenishment alarm and sends it to the warehouse management personnel terminal via SMS.

[0020] The data analysis and reporting module is used to analyze inventory data, identify equipment usage trends, and predict future demand.

[0021] The inbound / outbound record module and the spare parts inventory alarm module utilize RFID smart tags for real-time inventory tracking and automatic replenishment notifications. All spare parts are equipped with RFID tags, each storing relevant data including spare part type, manufacturer, model, shelf, and project. The RFID reader connects to the secondary equipment intelligent management system, processes data from the tags, and updates inventory changes in real time. The secondary equipment intelligent management system verifies the shelf information recorded on the spare parts tags to ensure consistency between the shelf and the spare parts type. If inconsistency is found, an error is reported, displaying an RFID data error. An alarm is also sent to the warehouse management personnel's terminal via SMS when items are placed in the wrong location.

[0022] The intelligent management system for secondary equipment includes:

[0023] The environmental monitoring status and alarm recording module is used to monitor the real-time temperature and humidity values ​​of the warehouse, as well as the camera surveillance footage, and to query the historical records and alarm records of the sensors and cameras. The sensors include temperature sensors, humidity sensors, and smoke detectors. All sensors are connected to the environmental monitoring status and alarm recording module to continuously collect and analyze environmental data, define safety thresholds for environmental temperature, humidity, and smoke conditions, and trigger an alarm when the corresponding thresholds are exceeded, automatically notifying the warehouse management personnel.

[0024] The DBO algorithm, optimized by dung beetles, is used to dynamically adjust the safety thresholds for temperature, humidity, and smoke, including:

[0025] 1.1: Determine the search space for the alarm threshold, i.e., the possible range of values ​​for the threshold;

[0026] 1.2: Select as the initial population size;

[0027] 1.3: For each dung beetle individual, calculate the false negative rate and false positive rate based on the threshold it represents, and then calculate the fitness function:

[0028] Fitness = False Negative Rate (Number of False Negatives / Total Number of Alarms) + False Alarm Rate (Number of False Alarms / Total Number of Alarms)

[0029] The goal is to minimize the fitness function, i.e., to reduce false negatives and false positives;

[0030] 1.4: Simulating the social behavior of dung beetles, i.e., individuals tend to follow individuals with lower fitness;

[0031] 1.5: Simulate the foraging behavior of dung beetles and search the solution space through random perturbation in the hope of finding a better threshold solution;

[0032] 1.6: Select dung beetle individuals for reproduction based on fitness to generate a new offspring population. Use Gaussian perturbation, with the perturbation magnitude based on a percentage of the threshold.

[0033] 1.7: Combine the offspring of the offspring with the current population to renew the population;

[0034] 1.8: Repeat steps 1.3 to 1.7, setting the maximum number of iterations to 100. If the improvement of the fitness function is less than 0.01% in 10 consecutive iterations, stop the algorithm.

[0035] 1.9: The threshold represented by the dung beetle individual with the lowest output fitness is taken as the optimal solution. After a total of 5 events of correct alarm, false alarm, and missed alarm, the threshold is re-examined to optimize the threshold.

[0036] The camera is equipped with an image recognition module for continuous real-time monitoring, covering the entire area of ​​the warehouse. The camera executes an AI video image recognition algorithm and is connected to the environmental monitoring status and alarm recording module.

[0037] The image recognition module includes a flame detection module and a personnel detection module. The flame detection module immediately issues an alarm upon detecting any signs of flame, triggers intelligent video recording and capture, and automatically notifies warehouse management personnel by telephone. The flame detection uses the YOLO target detection algorithm, which boasts high accuracy and stability. The personnel detection module automatically shuts off all warehouse lights if no one is detected in any area for 10 minutes.

[0038] The secondary equipment warehouse uses a constant temperature and humidity precision air conditioner and is equipped with an electrostatic eliminator and an ion fan to reduce the generation of static electricity in the warehouse.

[0039] The secondary equipment warehouse is equipped with water-blocking dams, drainage ditches and floor drains around the ground to prevent the transfer and seepage of groundwater. Water-sensitive detection instruments or components are installed around the warehouse's potential leakage areas, including under the warehouse floor and around the air conditioners, to detect and alarm on the warehouse's waterproofing.

[0040] The beneficial effects of this invention are that, compared with the prior art, this invention provides an intelligent warehouse system for secondary equipment in substations, which can ensure the normal operation of the warehouse and the safety and reliability of the equipment. In the event of a sudden failure or emergency, the warehouse can quickly provide the necessary equipment and accessories to help the substation quickly resume operation and reduce the impact of the accident on the power grid and users.

[0041] Alarm strategies are optimized based on Deep Q-Networks (DQNs). DQNs can handle high-dimensional and continuous state spaces, making them well-suited for complex environments such as intelligent warehouse monitoring systems, which involve large amounts of sensor readings and historical data. DQN models can continuously learn from new data, constantly improving alarm strategies to adapt to changes in the secondary equipment warehouse environment, thereby enhancing the practicality and accuracy of the alarm strategies.

[0042] Based on historical data and real-time changes in warehouse environmental parameters, the Dung Beetle Optimization Algorithm (DBO) dynamically adjusts the safety thresholds for temperature, humidity, and smoke. By simulating dung beetle behavior, the DBO algorithm possesses strong global search capabilities, helping to find optimal threshold settings and improve the accuracy of the alarm system. Even in complex and ever-changing monitoring environments, DBO maintains good search efficiency and stability, reducing false alarms and missed alarms. The DBO algorithm is simple in principle, has few parameters, and is easy to implement and apply in existing alarm systems, requiring no complex mathematical models or large computational resources. By simulating the random walking mechanism of dung beetles, DBO effectively escapes local optima, preventing threshold adjustments from getting trapped in local extrema and improving the comprehensiveness and accuracy of threshold adjustments.

[0043] Intelligent flame detection based on the YOLO algorithm: The YOLO model can process images at extremely high speeds, achieving real-time detection. This is particularly important for flame detection, as flames can appear suddenly and require a rapid response. Many pre-trained YOLO models are available for free download and application, accelerating the development process and reducing the need for data collection. With proper training and optimization, the YOLO model can achieve high detection accuracy, ensuring accurate flame identification. The YOLO model can adapt to different environmental conditions, including varying lighting conditions and background noise, making it suitable for applications in secondary equipment warehouses. Attached Figure Description

[0044] Figure 1 This is a schematic diagram showing the regional division of the intelligent warehouse for secondary equipment in substations according to the present invention.

[0045] Figure 2 This is a schematic diagram of the functional pages of the intelligent management system for secondary equipment in this invention.

[0046] Figure 3 This is a schematic diagram of the environmental monitoring and intelligent alarm system in this invention.

[0047] Figure 4 This is a flowchart of the safety threshold optimization using the dung beetle optimization algorithm in this invention. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0049] The embodiments described in this application are merely some, not all, embodiments of the present invention. Based on the spirit of the present invention, other embodiments obtained by those skilled in the art without inventive effort are all within the protection scope of the present invention.

[0050] To address the shortcomings of existing technologies, this invention proposes an intelligent warehouse system for substation secondary equipment. This system enables the warehouse to rapidly provide necessary equipment and spare parts in the event of sudden faults or emergencies, helping the substation quickly resume operation and reducing the impact of accidents on the power grid and users. Secondary equipment warehouses have their own characteristics and requirements in terms of equipment storage, security protection, management, and maintenance, necessitating a series of measures to ensure the normal operation of the warehouse and the safety and reliability of the equipment. Secondary equipment warehouses typically store various types of electrical equipment, such as relays, protection devices, and switchgear. These devices have special requirements for storage and management, needing to be kept dry, well-ventilated, and protected from environmental factors such as humidity and dust. Secondary equipment is often critical to substation operation, therefore its safety and integrity are paramount. Secondary equipment warehouses need to be equipped with security facilities, such as surveillance cameras and intrusion alarm systems, to prevent unauthorized personnel from entering or damaging the equipment. Due to the importance of secondary equipment, strict management and traceability are necessary. The warehouse needs to establish a comprehensive equipment management system, including functions such as equipment entry registration, exit records, and equipment status tracking, to ensure that the use and maintenance of the equipment are effectively managed and monitored. To ensure the normal operation and safety of secondary equipment, the warehouse needs to be inspected and maintained regularly. This includes regular inspections, cleaning, maintenance, and necessary repairs and replacements to ensure equipment performance and reliability. To cope with unexpected failures and equipment damage, the secondary equipment warehouse needs to establish a spare parts management system and develop corresponding replenishment plans. Warehouse managers need to rationally arrange the storage and replenishment of spare parts based on equipment usage and maintenance needs to ensure timely replacement and repair of equipment when needed. Based on the above requirements, an intelligent warehouse system for substation secondary equipment was developed.

[0051] The area division method of the intelligent warehouse for substation secondary equipment disclosed in this invention is as follows: Figure 1As shown, the warehouse is divided into eight zones: management zone, common tools and equipment zone, testing instrument zone, spare parts zone, relay protection spare parts zone, automation spare parts zone, and network security spare parts zone. Each storage area is labeled using category names, including visual aids such as symbols and color coding, with different areas marked with different colors. Physical dividing lines are used between different categories to delineate the end of one category and the beginning of another. The management zone contains warehouse management regulations and safety precautions, houses office computers and servers, and deploys a secondary equipment intelligent management system. The warehouse management regulations include: personnel responsibilities and obligations, spare parts entry and exit management regulations, warehouse safety management regulations, warehouse emergency response plans, and daily patrol regulations. Patrol route maps with arrows are posted on the ground for guidance.

[0052] Secondary equipment is typically critical to the operation of a substation, therefore its safety and integrity are paramount. Secondary equipment storage facilities must be equipped with security measures to prevent unauthorized personnel from entering or damaging the equipment.

[0053] First, authorized personnel are pre-registered in the secondary equipment intelligent management system, and their facial data is stored. A facial recognition scanner is installed at the warehouse entrance, integrated with the electronic door lock. Facial recognition technology is used for detection, allowing only authorized personnel to enter the warehouse. Managers manage and monitor access permissions by setting up access control software. Access levels for different groups or individuals are defined. Personnel entry and exit are recorded.

[0054] The functional pages of the secondary equipment intelligent management system are as follows: Figure 2 As shown, it includes a spare parts search and query module, an inbound and outbound record module, a spare parts inventory alarm record module, an environmental monitoring status and alarm record module, a user management module, and a warehouse management system query module.

[0055] The spare parts search module is used to search for spare parts based on names and other information, and displays spare parts information and their corresponding shelves based on the search results. The inbound / outbound record module is used for inventory management, allowing users to view equipment inbound and outbound times. The spare parts inventory alarm module generates equipment inventory alarm records based on equipment inventory quantities. The data analysis and reporting module uses inventory data for analysis, such as identifying equipment usage trends, predicting future demand, and optimizing inventory levels. The environmental monitoring status and alarm record module monitors real-time temperature, humidity, and camera surveillance footage in the warehouse, and allows users to query sensor and camera historical records and alarm records. The user management module provides the username and password required for warehouse management personnel to log in to the secondary equipment intelligent management system, as well as the terminal numbers of warehouse management and security personnel. The warehouse management system query includes queries for personnel responsibilities and obligations, spare parts inbound / outbound management systems, warehouse security management systems, warehouse emergency response plans, and daily patrol systems.

[0056] In the inventory management process of the aforementioned inbound / outbound record module, smart tags (RFID) are used for real-time inventory tracking and automatic replenishment notifications. Specifically, all spare parts are equipped with RFID tags. Each tag stores relevant data about the spare part, including spare part type, manufacturer, model, shelf, and item. RFID readers are installed at the warehouse entrance to capture data from the tags, record inbound and outbound information, and store the records. The RFID readers are connected to the secondary equipment intelligent management system. This system processes the data from the tags and updates inventory changes in real time. The management system verifies the shelf information recorded on the spare part tags to ensure consistency between the shelf and the spare part type. If there is a discrepancy, an error is reported, displaying an RFID data error. An alarm is issued to the warehouse manager when an item is placed in the wrong location, and a text message is sent to the warehouse manager's terminal. A minimum inventory level replenishment threshold is defined for each spare part. When the inventory falls below this threshold, the system triggers a replenishment alarm, which is also sent to the warehouse manager's terminal via text message. This system will improve inventory accuracy, reduce manual work, and ensure timely replenishment of spare parts.

[0057] The secondary equipment warehouse is equipped with sensors and cameras for temperature and humidity detection, smoke detection, personnel detection, and flame detection. This information is transmitted to the environmental monitoring status and alarm recording module, which automatically notifies warehouse management or security personnel when an anomaly is detected. The system structure is as follows: Figure 3 As shown.

[0058] The sensors include a temperature sensor, a humidity sensor, and a smoke detector. All sensors are connected to an environmental monitoring status and alarm recording module to continuously collect and analyze environmental data. Safety thresholds are defined for ambient temperature, humidity, and smoke levels. When these thresholds are exceeded, the system triggers an alarm and automatically notifies warehouse management or security personnel by telephone.

[0059] In a preferred embodiment, the safety thresholds for temperature, humidity, and smoke are dynamically adjusted based on historical data and real-time changes in the warehouse's environmental parameters. The Dung Beetle Optimization (DBO) algorithm, by simulating dung beetle behavior, possesses strong global search capabilities, helping to find optimal threshold settings and improve the accuracy of the alarm system. DBO maintains good search efficiency and stability even in complex and changing monitoring environments, reducing false alarms and missed alarms. The DBO algorithm is simple in principle, has few parameters, and is easy to implement and apply in existing alarm systems, requiring no complex mathematical models or large computational resources. By simulating the random walking mechanism of dung beetles, DBO can effectively escape local optima, preventing threshold adjustments from getting trapped in local extrema and improving the comprehensiveness and accuracy of threshold adjustments.

[0060] To optimize alarm thresholds using the dung beetle optimization algorithm, follow these steps: Figure 4 As shown.

[0061] 1.1 Determine the search space for the alarm threshold, that is, the possible range of values ​​for the threshold.

[0062] 1.2 Initialize the population by selecting 30 as the initial population size to balance the breadth and depth of the search.

[0063] 1.3 For each dung beetle individual, calculate the false negative rate and false positive rate based on the threshold it represents, and then calculate the fitness function:

[0064] Fitness = False Negative Rate (Number of False Negatives / Total Number of Alarms) + False Alarm Rate (Number of False Alarms / Total Number of Alarms)

[0065] The goal is to minimize the fitness function, i.e., to reduce false negatives and false positives.

[0066] 1.4 Simulate the social behavior of dung beetles, i.e., individuals tend to follow individuals with lower fitness (i.e., the sum of false negative and false positive rates).

[0067] 1.5 Simulate the foraging behavior of dung beetles and search the solution space through random perturbation in the hope of finding a better threshold solution.

[0068] 1.6: Select dung beetle individuals for reproduction based on fitness to generate a new offspring population. Use fitness ratio selection to ensure that superior threshold solutions have a higher chance of reproduction. Implement an elitist strategy to guarantee that the top 10% of optimal solutions are directly inherited by the next generation. Employ Gaussian perturbation, with the perturbation size based on a percentage of the threshold, such as 5% of the threshold, to avoid getting trapped in local optima.

[0069] 1.7: Combine the offspring produced from the reproduction with the current population and select as needed to renew the population.

[0070] 1.8: Repeat steps 1.3 to 1.7, setting the maximum number of iterations to 100. If the improvement of the fitness function is less than 0.01% in 10 consecutive iterations, stop the algorithm to prevent premature convergence or excessive iteration.

[0071] 1.9: The threshold represented by the dung beetle individual with the lowest output fitness (i.e., the minimum sum of false alarm and false alarm rates) is taken as the optimal solution. After every 5 occurrences of relevant events (including correct alarms, false alarms, and false alarms), the threshold is re-examined to optimize it, ensuring the reasonableness of the threshold and the effectiveness of the alarm system.

[0072] The environmental monitoring status and alarm recording module further utilizes reinforcement learning algorithms to optimize the alarm strategy. Deep Q-Network (DQN) has significant advantages in optimizing alarm strategies, especially in handling complex, dynamically changing environments and decision-making scenarios that require consideration of long-term impacts. The specific implementation process of using Deep Q-Network (DQN) to optimize the alarm strategy includes:

[0073] 2.1: Collect and preprocess warehouse monitoring data, including sensor readings (temperature, humidity, smoke levels, etc.), equipment status, and historical alarm records. Convert this data into feature vectors, which serve as the input states s of the DQN.

[0074] 2.2: List all possible actions the agent can take, including issuing an alarm, not issuing an alarm, turning off the power, adjusting sensor thresholds, etc. Assign a unique index to each action.

[0075] 2.3: Design the reward function based on business requirements. The reward function encompasses multiple aspects to ensure system security and efficiency. The following factors should be considered when designing the reward function:

[0076] 1) Alarm accuracy: Reduce false alarms and missed alarms.

[0077] R accuracy =-(w fp ·FP+w fn ·FN)

[0078] Where FP represents the number of false alarms, FN represents the number of false negatives, and w fp and w fn These are the corresponding weighting coefficients.

[0079] 2) Response time: Real alarms that encourage rapid response.

[0080] R response =w time (Alarm time - Response time)

[0081] w time These are the corresponding weighting coefficients.

[0082] 3) Alarm handling efficiency: Effective alarm handling measures are encouraged, and the success rate is the ratio of successfully handled alarms to total alarms.

[0083] R efficiency =w eff • Processing success rate

[0084] Where w eff These are the corresponding weighting coefficients.

[0085] 4) User satisfaction: Consider the satisfaction of warehouse management personnel with the alarm system, with a satisfaction score of 1-3 points, and 3 points being the most satisfactory.

[0086] R satisfaction =w sat Satisfaction rating

[0087] Where w sat These are the corresponding weighting coefficients.

[0088] The final reward function is a weighted sum of the above awards:

[0089] r = R accuracy +R response +R efficiency +R satisfaction

[0090] Based on expert experience and the actual conditions of the warehouse, the weight value w fp w fn w time w eff w sat Take values ​​of 0.3, 0.3, 0.1, 0.2, and 0.1 respectively, and adjust them according to the actual situation.

[0091] 2.4: A deep neural network is used to approximate the Q-value function. This network has one input layer, two hidden layers (with ReLU as the activation function), and one output layer. It employs the Adam optimizer with a learning rate α of 0.001 and a decay rate of 0.99. Mean squared error loss (MSE Loss) with L2 regularization and a weight decay coefficient of 0.001 is used to prevent overfitting. A greedy ε-policy is employed, with an initial ε value of 1, decreasing by 0.01 every 1000 training steps until a minimum value of 0.1 is reached. The network accepts states as input and outputs the Q-value for each possible action.

[0092] 2.5: During training, all experienced states, actions, rewards, and the next state are stored in an experience pool. In each training step, a mini-batch of samples is randomly drawn from the experience pool for training.

[0093] 2.6: A separate target network is used to compute the target Q-value, which copies weights from the online network at regular intervals. The target network is used to compute the target Q-value to reduce variance during training.

[0094] 2.7: Define the Mean Squared Error (MSE) loss function as follows:

[0095]

[0096] Where r is the immediate reward, γ is the discount factor used to weigh the importance of immediate reward and future reward, and is set to 0.95. Q'(s',a') is the Q value of the target network outputting the next state s' and action a', and Q(s,a) is the Q value of the online network outputting the current state s and action a.

[0097] 2.8: Use the gradient descent algorithm (Adam optimizer) to update the network weights in order to minimize the loss function.

[0098] In each training step, the weights of the online network are updated and periodically copied to the target network.

[0099] Update Q value:

[0100] Q(s,a)←Q(s,a)+α[r+γmax a' Q'(s',a')-Q(s,a)]

[0101] Where α is the learning rate and γ is the discount factor.

[0102] 2.9: An ε-greedy strategy is employed to balance exploration (randomly selecting actions) and exploitation (selecting the best action based on the current network conditions). As training progresses, the ε value is gradually reduced to decrease random exploration.

[0103] 2.10: Use a separate test set to evaluate the model's performance. If the model's performance meets the requirements, deploy it to a real-world environment. Optimize the alerting strategy through the above steps to improve system efficiency and security.

[0104] The cameras are installed inside the warehouse. In a preferred embodiment, six cameras equipped with image recognition modules are installed for 24 / 7 real-time monitoring, covering the entire warehouse area. Utilizing advanced AI video image recognition algorithms, the image recognition modules are divided into flame detection and personnel detection modules. The cameras are connected to an environmental monitoring status and alarm recording module.

[0105] The flame detection module uses the YOLO model to achieve intelligent flame detection.

[0106] The flame detection module uses a publicly available fire and flame detection dataset. All images are resized to 416×416 pixels and standardized. The YOLOv8 architecture is chosen, as it strikes a good balance between speed and accuracy. The YOLO loss function is used, with a learning rate of 0.001 and an SGD optimizer. The model is trained on the flame detection dataset, using early stopping to prevent overfitting. The model is evaluated on a test set, using mAP (mean Average Precision) as the performance metric. Based on the performance on the test set, the optimal detection threshold is determined. The trained model is then deployed to a real-time monitoring system.

[0107] The trained YOLO model is integrated into a real-time monitoring system to perform real-time flame detection on images captured by the camera. The initial image processing includes:

[0108] High-sensitivity cameras are used to capture video streams from inside the warehouse. The cameras should have good low-light performance to ensure reliable flame detection under varying lighting conditions.

[0109] Due to the spectral characteristics of flames, images within a specific wavelength range can be used to enhance flame features. Image preprocessing includes steps such as grayscale conversion and histogram equalization.

[0110] I'(x,y)=T·I(x,y)

[0111] Where I'(x,y) is the processed image intensity, I(x,y) is the original image intensity, and T is the transformation matrix designed based on the flame spectral characteristics.

[0112] After image processing, the image is input into the YOLO model for flame recognition, ensuring the effectiveness of flame recognition.

[0113] For the flame detection module, once a sign of flame is detected, the system will immediately issue an alarm after 3 seconds, triggering intelligent video recording and capture, and automatically notifying warehouse managers and security personnel by telephone, reminding relevant personnel to take timely action, thereby effectively preventing the occurrence of a fire.

[0114] The personnel detection module will automatically turn off all warehouse lights if it detects that no one is in any area of ​​the warehouse for 10 minutes, thus saving energy.

[0115] For physical safety, a dedicated temperature and humidity control precision air conditioner is used to ensure constant temperature and humidity in all areas. The warehouse is equipped with static eliminators and ion fans to reduce static electricity generation. Static-free wristbands are placed at the warehouse entrance for staff to wear when operating electronic equipment, reducing the risk of static electricity. Anti-static shoe covers are also provided at the entrance for staff to wear upon entry. Water barriers, drainage ditches, and floor drains are installed around potential leakage areas to prevent the transfer and seepage of groundwater caused by ruptured air conditioning drainage pipes. Water-sensitive detection instruments or components, such as leak detection ropes, are installed around potential leakage areas, such as under the warehouse floor and around the air conditioning units, to perform waterproofing checks and trigger alarms.

[0116] The beneficial effects of this invention are that, compared with the prior art, this invention provides an intelligent warehouse system for secondary equipment in substations, which can ensure the normal operation of the warehouse and the safety and reliability of the equipment. In the event of a sudden failure or emergency, the warehouse can quickly provide the necessary equipment and accessories to help the substation quickly resume operation and reduce the impact of the accident on the power grid and users.

[0117] Alarm strategies are optimized based on Deep Q-Networks (DQNs). DQNs can handle high-dimensional and continuous state spaces, making them well-suited for complex environments such as intelligent warehouse monitoring systems, which involve large amounts of sensor readings and historical data. DQN models can continuously learn from new data, constantly improving alarm strategies to adapt to changes in the secondary equipment warehouse environment, thereby enhancing the practicality and accuracy of the alarm strategies.

[0118] Based on historical data and real-time changes in warehouse environmental parameters, the Dung Beetle Optimization Algorithm (DBO) dynamically adjusts the safety thresholds for temperature, humidity, and smoke. By simulating dung beetle behavior, the DBO algorithm possesses strong global search capabilities, helping to find optimal threshold settings and improve the accuracy of the alarm system. Even in complex and ever-changing monitoring environments, DBO maintains good search efficiency and stability, reducing false alarms and missed alarms. The DBO algorithm is simple in principle, has few parameters, and is easy to implement and apply in existing alarm systems, requiring no complex mathematical models or large computational resources. By simulating the random walking mechanism of dung beetles, DBO effectively escapes local optima, preventing threshold adjustments from getting trapped in local extrema and improving the comprehensiveness and accuracy of threshold adjustments.

[0119] Intelligent flame detection based on the YOLO algorithm: The YOLO model can process images at extremely high speeds, achieving real-time detection. This is particularly important for flame detection, as flames can appear suddenly and require a rapid response. Many pre-trained YOLO models are available for free download and application, accelerating the development process and reducing the need for data collection. With proper training and optimization, the YOLO model can achieve high detection accuracy, ensuring accurate flame identification. The YOLO model can adapt to different environmental conditions, including varying lighting conditions and background noise, making it suitable for applications in secondary equipment warehouses.

[0120] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A substation secondary equipment intelligent warehouse system, comprising a secondary equipment warehouse and a secondary equipment intelligent management system, characterized in that: The secondary equipment warehouse is divided into a management area, a common tools area, a testing instrument area, a spare parts area, a relay protection spare parts area, an automation spare parts area, and a network security spare parts area. The warehouse is equipped with sensors and cameras for temperature and humidity detection, smoke detection, personnel detection, and flame detection. It also features RFID equipment to capture RFID tag data and record inbound and outbound information. Furthermore, it has facial recognition scanners for authorized personnel authentication and access control. The secondary equipment intelligent management system is used to query spare parts, display spare parts information, record and manage the entry and exit of spare parts, issue inventory alarms based on the quantity of spare parts, analyze inventory data, monitor and query warehouse environment data, and manage warehouse management personnel. The flame detection adopts the YOLOv8 architecture and uses the YOLO loss function. First, the YOLO model is trained on the flame detection dataset. Early stopping technique is used to prevent overfitting. The model is evaluated on the test set and mAP is used as the performance metric. Based on the performance on the test set, the optimal detection threshold is determined and the trained model is deployed to the real-time monitoring system. Real-time flame detection is performed on images captured by the camera. The first step is image processing, including: The video stream inside the warehouse was captured using a camera, and images within a specific wavelength range were used to enhance the characteristics of the flames. Image preprocessing included grayscale conversion and histogram equalization. I'(x,y)=T·I(x,y) Where I'(x,y) is the processed image intensity, I(x,y) is the original image intensity, and T is the transformation matrix based on the flame spectral characteristics; The processed image is then input into the trained YOLO model for flame detection. The intelligent management system for secondary equipment includes: The environmental monitoring status and alarm recording module is used to monitor the real-time temperature and humidity values ​​of the warehouse and the camera surveillance footage, and to query the historical records and alarm records of the sensors and cameras. The sensors include temperature sensors, humidity sensors and smoke detectors. All sensors are connected to the environmental monitoring status and alarm recording module to continuously collect and analyze environmental data, define safety thresholds for environmental temperature, humidity and smoke status, and trigger an alarm when the corresponding thresholds are exceeded, automatically notifying the warehouse management personnel. The DBO algorithm, optimized by dung beetles, is used to dynamically adjust the safety thresholds for temperature, humidity, and smoke, including: 1.1: Determine the search space for the alarm threshold, i.e., the possible range of values ​​for the threshold; 1.2: Select 30 as the initial population size; 1.3: For each dung beetle individual, calculate the false negative rate and false positive rate based on the threshold it represents, and then calculate the fitness function: Fitness = False negative rate + False positive rate The false alarm rate is calculated as: (Number of false alarms not detected / Total number of alarms) The false alarm rate is calculated as: (Number of false alarms not detected / Total number of alarms) The goal is to minimize the fitness function, i.e., to reduce false negatives and false positives; 1.4: Simulating the social behavior of dung beetles, i.e., individuals tend to follow individuals with lower fitness; 1.5: Simulate the foraging behavior of dung beetles and search the solution space through random perturbation in the hope of finding a better threshold solution; 1.6: Select dung beetle individuals for reproduction based on fitness to generate a new offspring population. Use Gaussian perturbation, with the perturbation magnitude based on a percentage of the threshold. 1.7: Combine the offspring of the offspring with the current population to renew the population; 1.8: Repeat steps 1.3 to 1.7, setting the maximum number of iterations to 100. If the improvement of the fitness function is less than 0.01% in 10 consecutive iterations, stop the algorithm. 1.9: The threshold represented by the dung beetle individual with the lowest output fitness is taken as the optimal solution. After a total of 5 events of correct alarm, false alarm and missed alarm, the threshold is re-examined to optimize the threshold.

2. The intelligent warehouse system for substation secondary equipment according to claim 1, characterized in that, Each storage area in the management area, common tools area, testing instrument area, accessories area, relay protection spare parts area, automation spare parts area, and network security spare parts area is labeled with a category name. Different areas are marked with different colors, and physical dividing lines are used to separate different categories. The management area is equipped with office computers and servers, and the intelligent management system for secondary equipment is deployed there.

3. The intelligent warehouse system for substation secondary equipment according to claim 2, characterized in that, The secondary equipment intelligent management system will pre-register the facial data of authorized personnel. The facial recognition scanner is integrated with the electronic door lock and uses facial recognition technology for detection. Only authorized personnel can enter the warehouse. Access permissions are managed and monitored by setting access control software to define access levels for different groups or individuals.

4. The intelligent warehouse system for substation secondary equipment according to claim 3, characterized in that, The intelligent management system for secondary equipment includes: The spare parts search and query module is used to search for spare parts by name and display the spare parts information and the shelf they belong to based on the search results; The inbound / outbound record module is used to manage inventory and view the time of equipment entry and exit. The spare parts inventory alarm module is used to generate equipment inventory alarm records based on the equipment inventory quantity; it defines the minimum inventory level replenishment threshold for each spare part, and when the inventory is lower than the corresponding threshold, it triggers a replenishment alarm and sends it to the warehouse management personnel terminal via SMS. The data analysis and reporting module is used to analyze inventory data, identify equipment usage trends, and predict future demand.

5. The intelligent warehouse system for substation secondary equipment according to claim 4, characterized in that, The inbound / outbound record module and the spare parts inventory alarm module utilize RFID smart tags for real-time inventory tracking and automatic replenishment notifications. All spare parts are equipped with RFID tags, and each tag stores relevant data, including spare part type, manufacturer, model, shelf, and project. The RFID reader is connected to the secondary equipment intelligent management system to process the data from the tags and update inventory changes in real time. The secondary equipment intelligent management system verifies the shelf information recorded on the spare parts tags to ensure that the shelf and spare parts type are consistent. If they are inconsistent, an error is reported, displaying an RFID data error. An alarm is sent to the warehouse management personnel when items are placed in the wrong location, and an SMS message is sent to the warehouse management personnel's terminal.

6. The intelligent warehouse system for substation secondary equipment according to claim 5, characterized in that, The camera is equipped with an image recognition module for continuous real-time monitoring, covering the entire area of ​​the warehouse. The camera executes an AI video image recognition algorithm and is connected to the environmental monitoring status and alarm recording module.

7. The intelligent warehouse system for secondary equipment in substations according to claim 6, characterized in that, The image recognition module includes a flame detection module and a personnel detection module. The flame detection module is used to immediately issue an alarm when it detects signs of flame, trigger intelligent video recording and capture, and automatically notify warehouse management personnel by telephone. The personnel detection module is used to turn off all warehouse lights when it detects that no one is in any area of ​​the warehouse for 10 minutes.

8. The intelligent warehouse system for secondary equipment in substations according to claim 7, characterized in that, The secondary equipment warehouse uses a constant temperature and humidity precision air conditioner and is equipped with an electrostatic eliminator and an ion fan to reduce the generation of static electricity in the warehouse.

9. The intelligent warehouse system for secondary equipment in substations according to claim 8, characterized in that, The secondary equipment warehouse is equipped with water-blocking dams, drainage ditches and floor drains around the ground to prevent the transfer and seepage of groundwater. Water-sensitive detection instruments or components are installed around the warehouse's potential leakage areas, including under the warehouse floor and around the air conditioners, to detect and alarm on the warehouse's waterproofing.

Images