Intelligent processing method and device for abnormal power supply, equipment and storage medium

Abstract

The invention discloses an intelligent processing method and device for power supply abnormity, equipment and a storage medium. The method comprises the following steps: constructing a monitoring network for a current online monitoring power supply system, acquiring sensing data of each sensing device in a set time slot, and determining sensing mutual information for each sensing device; on the basis of a logarithmic distance path loss model, calculating path loss between each sensing device and the master station device in the set time slot, and calculating communication capacity between each sensing device and the master station device in the set time slot according to the Shannon second theorem; performing three-layer control preprocessing on the obtained sensing data, and predicting power supply data; and maximizing the success probability and minimizing the total energy consumption by using a meta-reinforcement learning algorithm, and optimizing to obtain an optimal control decision. Therefore, the processing speed and the computing power are improved, unnecessary noise data are filtered out, the processing accuracy is improved, the success rate of processing coping decisions is remarkably improved, and the reliability and the stability of a power supply system are improved.

Description

Technical Field

[0001] The present application relates to the technical field of integrated synergy computing and control, specifically to technical fields such as smart grid and meta-reinforcement learning, and especially to an intelligent processing method, device, equipment and storage medium for power supply anomalies. Background Art

[0002] Traditional abnormal power outage handling technology is centered around "layered protection, local action, and manual recovery." It uses multi-level relay protection devices to monitor grid parameters, identify fault types (such as short circuits and ground faults), and rely on section switches and automatic reclosing technology to isolate the faulty section to prevent the fault from spreading. At the same time, the backup power system completes switching to provide temporary power supply for critical loads.

[0003] However, traditional power anomaly handling methods are limited by communication delays, empirical constant setting and manual intervention efficiency. They have bottlenecks in complex fault scenarios and recovery speeds, and cannot immediately handle corresponding issues in the event of power fluctuations or instantaneous power outages. Especially in high-load or multi-tasking environments, it may not be possible to fully save all the data being processed. Traditional methods often rely on preset rules or hardware device triggers, lack the ability to adaptively adjust to different power anomaly scenarios, and are not flexible and accurate enough in responding to abnormal events in complex power environments. Summary of the Invention

[0004] The present application provides an intelligent processing method, device, equipment and storage medium for power supply anomalies to solve the problems that existing power supply anomaly processing solutions are highly dependent on manual scheduling and have large delays, and the response is not flexible and accurate enough.

[0005] The technical solution is as follows:

[0006] In a first aspect, a method for intelligently processing power supply anomalies is provided, comprising:

[0007] Constructing a monitoring network for the power supply system currently being monitored online, the monitoring network comprising a master station device and multiple sensing devices;

[0008] Acquire the sensing data of each sensing device in a set time slot and determine the sensing mutual information for each sensing device;

[0009] Based on the logarithmic distance path loss model, the path loss between each sensing device and the master station device in the set time slot is calculated, and the communication capacity between each sensing device and the master station device in the set time slot is calculated according to Shannon's second theorem;

[0010] Based on the path loss and communication capacity between each sensing device and the master station device, the acquired sensing data is pre-processed using three layers of control and the power supply data is predicted.

[0011] According to the success control probability and total energy consumption defined for the power supply system, a meta-reinforcement learning algorithm is used to maximize the success probability and minimize the total energy consumption, thereby optimizing and obtaining the optimal control decision.

[0012] In one possible implementation, acquiring sensing data of each sensing device in a set time slot and determining sensing mutual information for each sensing device specifically include:

[0013] Acquire sensing data of each sensing device in a set time slot, and determine, based on the sensing data, an association variable between each sensing device and each sensing target in a plurality of sensing targets;

[0014] Using the determined correlation variables, calculate the channel gain of each sensing device;

[0015] The perceptual mutual information of each sensing device is calculated based on the channel gain of each sensing device and the associated variables of the corresponding multiple sensing targets.

[0016] In one possible implementation, based on a logarithmic distance path loss model, calculating the path loss between each sensing device and the master station device in a set time slot n specifically includes:

[0017]

[0018] Where L(r0) represents the reference gain when the distance between the sth sensing device and the master device is r0 = 1 meter, μ is the path loss index, A variable representing shadow fading, The variable represents the path loss, and r(s, MD) represents the distance between the sth sensing device and the master station device.

[0019] In a possible implementation, the communication capacity between each sensing device and the master device in a set time slot n is calculated according to Shannon's second theorem, specifically including:

[0020]

[0021] Among them, W s,com [n] represents the communication bandwidth between the sth sensing device and the master device, P s,com represents the power used for communication between the sth sensing device and the master device, n MD The communication noise power of the master device.

[0022] In one possible implementation, based on the path loss and communication capacity between each sensing device and the master station device, three-layer control preprocessing is performed on the acquired sensing data to predict power supply data, specifically including:

[0023] Based on the path loss and communication capacity between each sensing device and the master device, the following are performed in sequence: calculating the energy consumption and delay of the first layer data locally on each sensing device, calculating the energy consumption and delay of the second layer data transmitted from each sensing device to the master device, and calculating the energy consumption and delay of the third layer data of the sensing data transmitted from the master device to each sensing device;

[0024] Calculating the total energy consumption of the power supply system using the first layer data energy consumption, the second layer data energy consumption, and the third layer data energy consumption; and calculating the total delay of the power supply system using the first layer data delay, the second layer data delay, and the third layer data delay;

[0025] The sensing data is filtered based on the total energy consumption and the total delay, and the power supply data is predicted using the filtered sensing data.

[0026] In one possible implementation, based on the success control probability and total energy consumption defined for the power supply system, a meta-reinforcement learning algorithm is used to maximize the success probability and minimize the total energy consumption to optimize and obtain the optimal control decision, specifically including:

[0027] The successful control probability formula is defined for the power supply system:

[0028]

[0029] Among them, ξ s [n]=1 or 0, when and When s [n]=1, otherwise, ξ s [n]=0,T th To calculate the delay threshold required for analyzing data, R th is the perceptual mutual information threshold, is the root mean square error between the predicted power supply data and the actual sensing data;

[0030] The power supply abnormality of the power supply system is represented by triples using the meta-reinforcement learning algorithm, where: represents the state space, represents the action space, represents the reward function, τ∈[0,1] is the discount factor, Specifically expressed as:

[0031]

[0032] in, It represents the additional reward obtained when the decision-making process is successfully completed in the set time slot n, otherwise it is punished with express;

[0033] By maximizing the success probability and minimizing the total energy consumption, the optimization obtains the best control decision.

[0034] In a second aspect, an intelligent processing device for power supply anomalies is provided, comprising:

[0035] A construction module is used to construct a monitoring network for the power supply system currently being monitored online, wherein the monitoring network includes a master station device and multiple sensing devices;

[0036] An acquisition module, configured to acquire the sensing data of each sensing device in a set time slot and determine the sensing mutual information for each sensing device;

[0037] a calculation module, configured to calculate the path loss between each sensing device and the master station device in a set time slot based on a logarithmic distance path loss model, and calculate the communication capacity between each sensing device and the master station device in the set time slot according to Shannon's second theorem;

[0038] A processing module is used to perform three-layer control preprocessing on the acquired sensing data based on the path loss and communication capacity between each sensing device and the master station device, and predict power supply data;

[0039] The optimization module is used to maximize the success probability and minimize the total energy consumption based on the success control probability and total energy consumption defined for the power supply system, and optimize the optimal control decision.

[0040] According to a third aspect, an electronic device is provided, including:

[0041] at least one processor; and

[0042] a memory communicatively connected to the at least one processor; wherein,

[0043] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of any possible implementation manner and the aspects described above.

[0044] In a fourth aspect, a computer-readable storage medium is provided, wherein the storage medium stores at least one instruction, and the at least one instruction is loaded and executed by a processor to implement the method of the above-mentioned aspect and any possible implementation manner.

[0045] In a fifth aspect, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the above-mentioned aspects and any possible implementation method.

[0046] The beneficial effects of the technical solution provided by this application include at least:

[0047] It can be seen from the above technical solution that the embodiment of the present application constructs a monitoring network for the power supply system currently monitored online, obtains the perception data of each sensing device in the set time slot, and determines the perception mutual information for each sensing device; based on the logarithmic distance path loss model, calculates the path loss between each sensing device and the master station device in the set time slot, and calculates the communication capacity between each sensing device and the master station device in the set time slot according to Shannon's second theorem; based on the path loss and communication capacity between each sensing device and the master station device, performs three-layer control preprocessing on the acquired perception data and predicts the power supply data; according to the successful control probability and total energy consumption defined for the power supply system, uses the meta-reinforcement learning algorithm to maximize the success probability and minimize the total energy consumption, and optimizes the best control decision. Thus, the processing speed and computing power are improved, and unnecessary noise data can be filtered out, the processing accuracy is improved, the success rate of processing and response decisions is significantly improved, and the overall reliability, stability and flexibility of the power supply system are improved.

[0048] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0050] Figure 1 This is a schematic diagram of the steps of the intelligent processing method for power supply anomalies provided in an embodiment of the present application.

[0051] Figure 2 This is a schematic diagram of the power supply system network architecture provided in an embodiment of the present application.

[0052] Figure 3 This is a structural block diagram of an intelligent processing device for power supply anomalies provided in yet another embodiment of the present application.

[0053] Figure 4This is a block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0054] The following description of exemplary embodiments of the present application is provided in conjunction with the accompanying drawings, which include various details of the embodiments of the present application to facilitate understanding. These details should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present application. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0055] Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0056] It should be noted that the terminal devices involved in the embodiments of the present application may include but are not limited to mobile phones, personal digital assistants (PDAs), wireless handheld devices, tablet computers and other smart devices; display devices may include but are not limited to personal computers, televisions and other devices with display functions.

[0057] In this document, the term "and / or" simply describes a relationship between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " in this document generally indicates that the related objects are in an "or" relationship.

[0058] In view of the fact that the existing power supply anomaly handling scheme is highly dependent on manual scheduling and has large delays, and the response is not flexible and accurate enough, this application proposes an intelligent handling scheme for power supply anomalies. The inventive concept of this scheme is: using the high mapping and collaborative work of the integrated architecture of synaesthesia and computing control and the meta-reinforcement learning algorithm to perform real-time perception and monitoring of the power supply system, achieving accurate perception and intelligent decision-making of power anomalies, optimizing data processing through the three-layer computing control architecture and the meta-reinforcement learning algorithm, filtering out energy consumption data from the perception data, thereby improving processing speed and computing power, and filtering out unnecessary noise data, improving processing accuracy, and significantly improving the success rate of processing and response decisions. At the same time, this scheme can not only dynamically adjust the control strategy of the processing response to ensure that the power supply equipment exits smoothly when the power is interrupted, avoiding data loss and hardware damage, but also has strong adaptability and robustness, improving the overall reliability and stability of the power supply system.

[0059] Reference Figure 1 , which is a schematic diagram of the steps of the intelligent processing method for power supply anomalies provided in an embodiment of the present application. The execution subject of the intelligent processing method can be an intelligent processing device for power supply anomalies. The intelligent processing device can be a hardware device or software module with computing, processing, storage and other functions, such as a computer, pad, smart phone, smart wearable device and other electronic devices, or a software module or component integrated and installed in such electronic devices.

[0060] Reference Figure 2 The power supply system network architecture diagram shown in the figure shows a power supply system 201 as the power supply network to be monitored. The power supply system 201 includes multiple sensing targets 1-k, which can be regarded as power supply equipment that provides different power supply requirements in the power supply system. Then, the power supply system 201 is monitored and controlled by an intelligent processing device 202 for power supply anomalies. The intelligent processing device 202 can be regarded as a monitoring network, including a master station device and multiple sensing devices 1-S. Each sensing device can obtain a set of sensing data about k sensing targets from the power supply system 201. In this way, a total of S groups of sensing data can be obtained, and each group of sensing data can include current, voltage, line loss, temperature and other data of these k sensing targets.

[0061] like Figure 1 As shown, the intelligent processing method for power supply anomaly may include the following steps:

[0062] Step 102: Construct a monitoring network for the power supply system currently being monitored online, wherein the monitoring network includes a master station device and multiple sensing devices.

[0063] In this application scheme, a power supply system for online monitoring can be constructed as follows: Figure 2 The monitoring network shown is used to cooperate with the power supply system to carry out effective and timely power supply abnormality response control.

[0064] Step 104: Acquire the sensing data of each sensing device in the set time slot, and determine the sensing mutual information for each sensing device.

[0065] Optionally, when obtaining the perception data of each perception device in a set time slot and determining the perception mutual information for each perception device, the perception data of each perception device in the set time slot can be obtained, and the association variables of each perception device and each perception target in the multiple perception targets can be determined based on the perception data; the channel gain of each perception device is calculated using the determined association variables; and the perception mutual information of each perception device is calculated based on the channel gain of each perception device and the association variables of the corresponding multiple perception targets.

[0066] Step 106: Based on the logarithmic distance path loss model, calculate the path loss between each sensing device and the master station device in the set time slot, and calculate the communication capacity between each sensing device and the master station device in the set time slot according to Shannon's second theorem.

[0067] In this application solution, the path loss between each sensing device and the master station device can be calculated using the following formula:

[0068]

[0069] Where L(r0) represents the reference gain when the distance between the sth sensing device and the master device is r0 = 1 meter, μ is the path loss index, A variable representing shadow fading, The variable represents the path loss, and r(s, MD) represents the distance between the sth sensing device and the master station device.

[0070] Furthermore, the communication capacity between each sensing device and the master device in the set time slot n is calculated according to Shannon's second theorem.

[0071]

[0072] Among them, W s,com [n] represents the communication bandwidth between the sth sensing device and the master device, P s,com represents the power used for communication between the sth sensing device and the master device, n MD The communication noise power of the master device.

[0073] Step 108: Based on the path loss and communication capacity between each sensing device and the master station device, perform three-layer control preprocessing on the acquired sensing data and predict the power supply data.

[0074] Optionally, when performing three-layer control preprocessing on the acquired perception data and predicting the power supply data based on the path loss and communication capacity between each perception device and the master station device, the following can be performed in sequence based on the path loss and communication capacity between each perception device and the master station device: calculating the first-layer data energy consumption and delay of each perception device locally, calculating the second-layer data energy consumption and delay transmitted by each perception device to the master station device, and calculating the third-layer data energy consumption and delay of the perception data transmitted by the master station device to each perception device; using the first-layer data energy consumption, the second-layer data energy consumption and the third-layer data energy consumption to calculate the total energy consumption of the power supply system; and using the first-layer data delay, the second-layer data delay and the third-layer data delay to calculate the total delay of the power supply system; filtering the perception data based on the total energy consumption and total delay, and using the filtered perception data to predict the power supply data.

[0075] Step 110: Based on the success control probability and total energy consumption defined for the power supply system, a meta-reinforcement learning algorithm is used to maximize the success probability and minimize the total energy consumption, thereby optimizing and obtaining the best control decision.

[0076] In the present application solution, first, a successful control probability formula can be defined for the power supply system:

[0077]

[0078] Among them, ξ s [n]=1 or 0, when and When s [n]=1, otherwise, ξ s [n]=0,T th To calculate the delay threshold required for analyzing data, R th is the perceptual mutual information threshold, is the root mean square error between the predicted power supply data and the actual sensing data;

[0079] Then, the power supply abnormality of the power supply system is represented as a triple using the meta-reinforcement learning algorithm, where represents the state space, represents the action space, represents the reward function, τ∈[0,1] is the discount factor, Specifically expressed as:

[0080]

[0081] in, It represents the additional reward obtained when the decision-making process is successfully completed in the set time slot n, otherwise it is punished with express;

[0082] Finally, the best control decision is obtained by optimizing the control strategy by maximizing the success probability and minimizing the total energy consumption.

[0083] Below, we take an integrated network architecture with telemetry, computing, and control as an example. This architecture is equipped with a master device (hereinafter referred to as MD) and a set of S sensing devices (hereinafter referred to as CD). These CDs use sensor signals to detect the current, voltage, line loss, temperature, and other characteristics of the power supply equipment to obtain sensor data on the power supply situation. The data is then preprocessed and offloaded to the MD via communication signals to meet the requirements of sensor data analysis and information fusion, further controlling the system's processing decisions, such as soft shutdown and delayed power off.

[0084] When the CD performs the sensing task, the sensor return signal received by the receiver (including information data such as current, voltage, line loss, and temperature) is converted into a set of sensing data. The amount of data (in bits) sensed by the CD s in time slot n can be modeled as:

[0085]

[0086] where α s represents the constant associated with the introduced data redundancy, v s Indicates the switching speed of the perception signal, N θ represents the number of quantized angles, f s represents the sampling frequency, δ s Indicates the number of quantization bits per sample.

[0087] Assume there are K sensing targets, define g sk [n]∈{0,1} represents the relationship between CD s and the perceived target If the associated variable between the two, if the perception target k is associated with CD s in time slot n, then g sk [n]=1, otherwise g sk [n] = 0. A CD can sense multiple targets in each time slot, but each target can only be sensed by one CD at most, that is, the following constraints are met

[0088] The CD transmits sensing signals such as those monitoring current, voltage, line loss, and temperature, and receives the return signal. The channel gain is expressed as:

[0089]

[0090] Where r(s,k) is the distance between CD s and the perceived target k, and β0 is the unit power when the distance r(s,k)=1.

[0091] In order to measure the performance of CD perception, the perceptual mutual information is calculated to quantify the information content between the detected target and the perception signal, so the perceptual mutual information is expressed as:

[0092]

[0093] in, is the signal-to-noise ratio of the perceived signal transmission, P s [n] is the power of the CD transmission perception signal, n s is the noise power, W s [n] is the perceived signal bandwidth.

[0094] In addition to perception, CD also needs to transmit pre-processed data to MD for further analysis and fusion.

[0095] First, according to the logarithmic distance path loss model, the path loss L between CDs and MD at time slot n is s,MD [n] is modeled as:

[0096]

[0097] Where L(r0) represents the reference gain when the distance between CD and MD is r0 = 1 meter, μ is the path loss exponent, A variable representing shadow fading, A variable representing path loss.

[0098] Secondly, according to Shannon's second theorem, the communication capacity between CDs and MD at time slot n is expressed as:

[0099]

[0100] Where W s,com [n] represents the communication bandwidth between CD and MD, P s,com represents the power used for communication between CDs and MD, n MD is the communication noise power.

[0101] Furthermore, to obtain accurate information, the MD subsequently needs to process the sensed data. On the one hand, the amount of data sensed by the CD can be large, and the CD is severely constrained in terms of size, energy, and computing power. On the other hand, the MD receiver continuously monitors the received signal to detect targets, but unwanted signals such as sensed echoes from other clutter, RF interference, and noise sources can obscure valuable data. The CD can eliminate these signals before offloading the sensed data to the MD. Therefore, a three-tier computing architecture is adopted to process the sensed data and predict power supply conditions.

[0102] In the first layer, CD preprocesses the sensory data through local computation to remove unnecessary signals. The energy consumption associated with this local data preprocessing is and delay It can be modeled as:

[0103]

[0104] in Represents the effective capacitance coefficient of the CDs local processor, which is a constant determined by the CD hardware specification. is the local processing density of CDs, represents the local computing power of CDs in time slot n.

[0105] The second layer, CD offloads the locally processed data to MD, which delays the offloading. and energy consumption It can be calculated as:

[0106]

[0107] where Φ s Represents the proportion of perception data output by CDs preprocessing.

[0108] The third layer, MD processes the measured data and outputs the energy consumption of control information and delay It can be given by the following formula:

[0109]

[0110] in represents the effective capacitance coefficient of the MD local processor, is the local processing density of MD, represents the local computing capacity of the MD in time slot n.

[0111] Then, the total energy consumption of the power supply system can be expressed as

[0112] At the same time, the predicted power supply data generated by the perception data processed by the three-layer computing architecture is defined as where d i is the weight coefficient, and the root mean square error used to evaluate the prediction performance

[0113] For the convenience of analysis, the system is considered successful when it meets the constraints of perception, communication, and computational control. Therefore, the success control probability of the system achieving soft shutdown and delayed power off is defined as:

[0114]

[0115] Among them, ξ s [n]=1 or 0, when and When s [n]=1, otherwise ξ s [n] = 0, where T th To calculate the delay threshold of the analyzed data, R th is the threshold for sensing the required data.

[0116] The success rate of soft shutdown and delayed power off can be maximized later. At the same time, minimize the energy consumption E[n] of the system to achieve efficient energy saving of the system, and abstract this problem into a Markov decision process using tuples Indicates that represents the state space, represents the action space, represents the reward function, τ∈[0,1] is the discount factor, Specifically expressed as:

[0117]

[0118] in Indicates the additional reward obtained when soft shutdown and delayed power off are achieved in time slot n, otherwise the penalty is express.

[0119] Define the state-action-value function of the meta-reinforcement learning algorithm as To evaluate the status Take action The value of , where π is the policy function, E(·) is the expected operation, for The next state of . Define the state value function as The value function measures the value of each state, that is, the future return that can be achieved in each state, where Indicates that in the meta-reinforcement learning algorithm, the parameter is ρ and the state is The following strategy is used to update the parameters using gradient descent: ∈ is the meta-learning rate. Then, the optimal control strategy is obtained through the defined objective function.

[0120] It should be noted that for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application.

[0121] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0122] Figure 3 FIG. 1 shows a structural block diagram of an intelligent processing device for power supply anomaly provided by an embodiment of the present application, as shown in FIG. Figure 3As shown. The intelligent processing device 300 for power supply anomalies in this embodiment may include a construction module 301, an acquisition module 302, a calculation module 303, a processing module 304 and an optimization module 305. Among them, the construction module 301 is used to build a monitoring network for the power supply system currently monitored online, and the monitoring network includes a master station device and multiple sensing devices. The acquisition module 302 is used to obtain the perception data of each sensing device in a set time slot, and determine the perception mutual information for each sensing device. The calculation module 303 is used to calculate the path loss between each sensing device and the master station device in the set time slot based on the logarithmic distance path loss model, and calculate the communication capacity between each sensing device and the master station device in the set time slot according to Shannon's second theorem. The processing module 304 is used to perform three-layer control preprocessing on the acquired perception data based on the path loss and communication capacity between each sensing device and the master station device, and predict the power supply data. The optimization module 305 is configured to optimize the optimal control decision by maximizing the success probability and minimizing the total energy consumption based on the success control probability and total energy consumption defined for the power supply system using a meta-reinforcement learning algorithm.

[0123] It should be noted that part or all of the intelligent processing device for power supply anomalies in this embodiment can be an application located in the local terminal, or it can also be a functional unit such as a plug-in or software development kit (SDK) set in the application located in the local terminal, or it can also be a processing engine located in the network side server, or it can also be a distributed system located on the network side. This embodiment does not specifically limit this.

[0124] It is understandable that the application may be a native program (nativeApp) installed on the local terminal, or may be a webpage program (webApp) of a browser on the local terminal, which is not limited in this embodiment.

[0125] Optionally, in a possible implementation of this embodiment, when the acquisition module 302 acquires the perception data of each perception device in a set time slot and determines the perception mutual information for each perception device, it is specifically used to acquire the perception data of each perception device in a set time slot, and determine the association variables of each perception device and each perception target in a plurality of perception targets based on the perception data; use the determined association variables to calculate the channel gain of each perception device; and calculate the perception mutual information of each perception device based on the channel gain of each perception device and the corresponding association variables of the plurality of perception targets.

[0126] Optionally, in a possible implementation of this embodiment, when the calculation module 303 calculates the path loss between each sensing device and the master station device in the set time slot n based on the logarithmic distance path loss model, it is specifically configured to calculate the path loss using the following formula:

[0127]

[0128] Where L(r0) represents the reference gain when the distance between the sth sensing device and the master device is r0 = 1 meter, μ is the path loss index, A variable representing shadow fading, The variable represents the path loss, and r(s, MD) represents the distance between the sth sensing device and the master station device.

[0129] Optionally, in a possible implementation of this embodiment, when calculating the communication capacity between each sensing device and the master station device in the set time slot n according to Shannon's Second Theorem, the calculation module 303 is specifically configured to calculate the communication capacity using the following formula:

[0130]

[0131] Among them, W s,com [n] represents the communication bandwidth between the sth sensing device and the master device, P s,com represents the power used for communication between the sth sensing device and the master device, n MD The communication noise power of the master device.

[0132] Optionally, in a possible implementation of this embodiment, the processing module 304 performs three-layer control preprocessing on the acquired perception data based on the path loss and communication capacity between each perception device and the master station device, and predicts the power supply data. Specifically, based on the path loss and communication capacity between each perception device and the master station device, the following steps are executed in sequence: calculating the first-layer data energy consumption and delay of each perception device locally, calculating the second-layer data energy consumption and delay transmitted by each perception device to the master station device, and calculating the third-layer data energy consumption and delay of the perception data transmitted by the master station device to each perception device; using the first-layer data energy consumption, the second-layer data energy consumption and the third-layer data energy consumption to calculate the total energy consumption of the power supply system; and using the first-layer data delay, the second-layer data delay and the third-layer data delay to calculate the total delay of the power supply system; filtering the perception data based on the total energy consumption and total delay, and using the filtered perception data to predict the power supply data.

[0133] Optionally, in a possible implementation of this embodiment, the optimization module 305 is specifically configured to define a successful control probability formula for the power supply system when optimizing to obtain the optimal control decision by using a meta-reinforcement learning algorithm to maximize the success probability and minimize the total energy consumption based on the successful control probability and total energy consumption defined for the power supply system:

[0134]

[0135] Among them, ξ s [n]=1 or 0, when and When s [n]=1, otherwise, ξ s [n]=0,T th To calculate the delay threshold required for analyzing data, R th is the perceptual mutual information threshold, is the root mean square error between the predicted power supply data and the actual perception data; the power supply anomaly of the power supply system is represented by a triple using the meta-reinforcement learning algorithm, where represents the state space, represents the action space, represents the reward function, τ∈[0,1] is the discount factor, Specifically expressed as:

[0136]

[0137] in, It represents the additional reward obtained when the decision-making process is successfully completed in the set time slot n, otherwise it is punished with Represents that the best control decision is obtained by maximizing the success probability and minimizing the total energy consumption.

[0138] In this embodiment, a monitoring network can be constructed for the power supply system currently being monitored online to obtain the perception data of each sensing device in a set time slot, and to determine the perception mutual information for each sensing device; based on the logarithmic distance path loss model, the path loss between each sensing device and the master station device in the set time slot is calculated, and the communication capacity between each sensing device and the master station device in the set time slot is calculated according to Shannon's second theorem; based on the path loss and communication capacity between each sensing device and the master station device, the acquired perception data is pre-processed in three layers of control and the power supply data is predicted; based on the success control probability and total energy consumption defined for the power supply system, the meta-reinforcement learning algorithm is used to maximize the success probability and minimize the total energy consumption, and the optimal control decision is optimized. Thus, the processing speed and computing power are improved, and unnecessary noise data can be filtered out, the processing accuracy is improved, the success rate of the processing response decision is significantly improved, and the overall reliability and stability of the power supply system are improved.

[0139] An embodiment of the present application provides a computer-readable storage medium, wherein the storage medium stores at least one instruction, and the at least one instruction is loaded and executed by a processor to implement the method for intelligent processing of power supply anomalies as described above.

[0140] An embodiment of the present application provides an electronic device, which includes a processor and a memory, wherein the memory stores at least one instruction, and the instruction is loaded and executed by the processor to implement the method for intelligent processing of power supply anomalies as described above.

[0141] In the technical solution of this application, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0142] Figure 4 A schematic block diagram of an example electronic device 400 that can be used to implement an embodiment of the present application is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application described and / or claimed herein.

[0143] like Figure 4As shown, electronic device 400 includes a computing unit 401, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 402 or a computer program loaded from a storage unit 408 into a random access memory (RAM) 403. Various programs and data required for the operation of electronic device 400 can also be stored in RAM 403. Computing unit 401, ROM 402, and RAM 403 are connected to each other via a bus 404. An input / output (I / O) interface 405 is also connected to bus 404.

[0144] Multiple components in the electronic device 400 are connected to the I / O interface 405, including an input unit 406, such as a keyboard, a mouse, etc.; an output unit 407, such as various types of displays, speakers, etc.; a storage unit 408, such as a magnetic disk, an optical disk, etc.; and a communication unit 409, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 409 allows the electronic device 400 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0145] The computing unit 401 can be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units that run machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 401 performs the various methods and processes described above, such as the method for intelligent processing of power supply anomalies. For example, in some embodiments, the method for intelligent processing of power supply anomalies can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as the storage unit 408. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 400 via the ROM 402 and / or the communication unit 409. When the computer program is loaded into the RAM 403 and executed by the computing unit 401, one or more steps of the method for intelligent processing of power supply anomalies described above can be performed. Alternatively, in other embodiments, the computing unit 401 may be configured in any other appropriate manner (for example, by means of firmware) to execute the method for intelligent processing of power supply anomalies.

[0146] Various embodiments of the systems and techniques described above can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, at least one input device, and at least one output device.

[0147] The program code for implementing the methods of the present application can be written in any combination of one or more programming languages. Such program code can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the program code is executed by the processor or controller, the functions / operations specified in the flow charts and / or block diagrams are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0148] In the context of the present application, a machine-readable medium can be a tangible medium that can contain or store a program for use by an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared or semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0149] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0150] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0151] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises through computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.

[0152] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this application can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this application can be achieved. This is not a limitation herein.

[0153] The above specific embodiments do not constitute a limitation on the scope of protection of this application. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application shall be included within the scope of protection of this application.

Claims

1. An intelligent processing method for power supply anomalies, characterized in that: include: Constructing a monitoring network for the power supply system currently being monitored online, the monitoring network comprising a master station device and multiple sensing devices; Acquire the sensing data of each sensing device in a set time slot and determine the sensing mutual information for each sensing device; Based on the logarithmic distance path loss model, the path loss between each sensing device and the master station device in the set time slot is calculated, and the communication capacity between each sensing device and the master station device in the set time slot is calculated according to Shannon's second theorem; Based on the path loss and communication capacity between each sensing device and the master station device, the acquired sensing data is pre-processed using three layers of control and the power supply data is predicted. According to the success control probability and total energy consumption defined for the power supply system, a meta-reinforcement learning algorithm is used to maximize the success probability and minimize the total energy consumption, thereby optimizing and obtaining the optimal control decision.

2. The method according to claim 1, wherein Acquire the sensing data of each sensing device in the set time slot and determine the sensing mutual information for each sensing device, specifically including: Acquire sensing data of each sensing device in a set time slot, and determine, based on the sensing data, an association variable between each sensing device and each sensing target in a plurality of sensing targets; Using the determined correlation variables, calculate the channel gain of each sensing device; The perceptual mutual information of each sensing device is calculated based on the channel gain of each sensing device and the associated variables of the corresponding multiple sensing targets.

3. The method according to claim 1, wherein Based on the logarithmic distance path loss model, the path loss between each sensing device and the master device in the set time slot n is calculated, specifically including: Where L(r0) represents the reference gain when the distance between the sth sensing device and the master device is r0 = 1 meter, μ is the path loss index, A variable representing shadow fading, The variable represents the path loss, and r(s, MD) represents the distance between the sth sensing device and the master station device.

4. The method according to claim 3, wherein The communication capacity between each sensing device and the master device in a set time slot n is calculated according to Shannon's second theorem, specifically including: Among them, W s,com [N] represents the communication bandwidth between the sth sensing device and the master device, P s,com represents the power used for communication between the sth sensing device and the master device, n MD The communication noise power of the master device.

5. The method according to claim 4, wherein Based on the path loss and communication capacity between each sensing device and the master station device, the three-layer control preprocessing is performed on the acquired sensing data to predict the power supply data, including: Based on the path loss and communication capacity between each sensing device and the master device, the following are performed in sequence: calculating the energy consumption and delay of the first layer data locally on each sensing device, calculating the energy consumption and delay of the second layer data transmitted from each sensing device to the master device, and calculating the energy consumption and delay of the third layer data of the sensing data transmitted from the master device to each sensing device; Calculating the total energy consumption of the power supply system using the first layer data energy consumption, the second layer data energy consumption, and the third layer data energy consumption; and calculating the total delay of the power supply system using the first layer data delay, the second layer data delay, and the third layer data delay; The sensing data is filtered based on the total energy consumption and the total delay, and the power supply data is predicted using the filtered sensing data.

6. The method according to claim 5, wherein According to the success control probability and total energy consumption defined for the power supply system, a meta-reinforcement learning algorithm is used to maximize the success probability and minimize the total energy consumption, thereby optimizing and obtaining the optimal control decision, specifically including: The successful control probability formula is defined for the power supply system: Among them, ξ s [n]=1 or 0, when and When s [n]=1, otherwise, ξ s [n]=0,T th To calculate the delay threshold required for analyzing data, R th is the perceptual mutual information threshold, is the root mean square error between the predicted power supply data and the actual sensing data; The power supply abnormality of the power supply system is represented by triples using the meta-reinforcement learning algorithm, where: represents the state space, represents the action space, represents the reward function, τ∈[0,1] is the discount factor, Specifically expressed as: in, It represents the additional reward obtained when the decision-making process is successfully completed in the set time slot n, otherwise it is punished with express; By maximizing the success probability and minimizing the total energy consumption, the optimization obtains the best control decision.

7. An intelligent processing device for power supply anomalies, characterized in that: include: A construction module is used to construct a monitoring network for the power supply system currently being monitored online, wherein the monitoring network includes a master station device and multiple sensing devices; An acquisition module, configured to acquire the sensing data of each sensing device in a set time slot and determine the sensing mutual information for each sensing device; a calculation module, configured to calculate the path loss between each sensing device and the master station device in a set time slot based on a logarithmic distance path loss model, and calculate the communication capacity between each sensing device and the master station device in the set time slot according to Shannon's second theorem; A processing module is used to perform three-layer control preprocessing on the acquired sensing data based on the path loss and communication capacity between each sensing device and the master station device, and predict power supply data; The optimization module is used to maximize the success probability and minimize the total energy consumption based on the success control probability and total energy consumption defined for the power supply system, and optimize the optimal control decision.

8. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 6.

10. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 6.

Images