Product life cycle management system and method based on smart power grid

Abstract

The invention relates to the technical field of smart power grids, and discloses a product life cycle management system and method based on a smart power grid, and the system comprises a data collection and compatibility management module which collects voltage, current and temperature data through a sensor network, converts the data into a structured data table with a timestamp, and stores the structured data table in a database; outputting the standardized data; and the edge computing and cloud collaborative computing module receives the standardized data, executes edge node data cleaning, de-noising preprocessing and threshold anomaly detection triggering, and uploads the cleaned data to the cloud to generate an energy efficiency optimization strategy. By integrating various sensors and an edge computing technology, real-time acquisition and rapid processing of multi-dimensional operation data of the power equipment are realized, and the system can rapidly respond to equipment state change and timely identify potential problems due to the efficient data processing capability, so that the equipment failure rate is reduced, and the operation reliability of the equipment is improved.

Description

Technical Field

[0001] The present invention relates to the technical fields of smart grid technology and product life cycle management technology, and specifically to a product life cycle management system and method based on a smart grid. Background Art

[0002] With the rapid development of smart grid technology, the management and maintenance of power equipment are facing increasingly complex challenges. Traditional product life cycle management methods rely on manual monitoring and manual data recording, unable to achieve real-time response and efficient decision-making. Such methods are inefficient and prone to delayed detection of equipment failures, increasing maintenance costs and downtime.

[0003] In this context, the integration of multiple sensors and the application of edge computing technology have received attention. By collecting multi-dimensional device operation data in real time, dynamic monitoring of device status is achieved, improving the reliability and availability of devices. However, existing systems have defects in data processing speed and anomaly detection capabilities and cannot meet the requirements of real-time response.

[0004] In addition, with the increasing severity of data privacy and security issues, traditional data storage methods cannot effectively protect user information and device data. The lack of a secure environment makes data face the risk of being tampered with or leaked during transmission and storage, bringing additional legal and economic burdens to enterprises.

[0005] In terms of resource management, after the end of the device life cycle, how to efficiently evaluate the recycling value and achieve resource reuse has become a major challenge in sustainable development. Traditional recycling management methods lack systematicness and scientific basis, leading to resource waste and failing to fully exploit the potential of resource reuse.

[0006] Therefore, those skilled in the art provide a product life cycle management system and method based on a smart grid to solve the problems raised in the above background art. Summary of the Invention

[0007] Aiming at the deficiencies of the prior art, the present invention provides a product life cycle management system and method based on a smart grid to solve the problems raised in the above background art.

[0008] To achieve the above objectives, the present invention is realized through the following technical solutions: A product life cycle management system based on a smart grid, the product life cycle management system includes:

[0009] A data acquisition and compatibility management module, which collects voltage, current, and temperature data through a sensor network, converts it into a structured data table with timestamps, and outputs standardized data;

[0010] The edge computing and cloud collaborative computing module receives standardized data, performs edge node data cleaning and denoising preprocessing, triggers threshold anomaly detection, and uploads the cleaned data to the cloud to generate an energy efficiency optimization strategy;

[0011] The intelligent fault prediction and preventive maintenance module constructs a device health model based on the generated energy efficiency optimization strategy, and outputs the remaining life curve, inspection cycle, and spare part replacement plan;

[0012] The data security module adds a device fingerprint to the standardized data, binds the fingerprint and data through the blockchain, processes the spare part replacement plan using the differential privacy mechanism, and outputs the encrypted data and the cloud storage path;

[0013] The resource recovery and reuse closed-loop management module, according to the remaining life curve and the encrypted storage path, parses the material list and working condition records when the device health threshold is triggered, and generates a disassembly plan and a material regeneration path;

[0014] The intelligent visualization and user interaction module integrates the disassembly plan, the material regeneration path, and the encrypted data, constructs a 3D topology map and a fault heat map, and pushes maintenance suggestions to the interactive dashboard.

[0015] Preferably, the data acquisition and compatibility management module further includes:

[0016] The multi-source heterogeneous data acquisition sub-module collects voltage V i (t), current I j (t), temperature T k (t) through the sensor network, and outputs the original data set D ij :

[0017] D ij ={(V i (t), I j (t), T k (t), t)},

[0018] where i, j, k represent the sensor numbers, and t is the acquisition timestamp;

[0019] The timestamp synchronization and alignment sub-module receives D ij output by the multi-source heterogeneous data acquisition sub-module, and aligns the multi-sensor timestamps through the time window sliding algorithm t ′ =t + τ i , where τ i is the clock offset of the i-th sensor, t ′ is the aligned unified timestamp, and generates the synchronized structured data table T ij (t ′ );

[0020] The structured data encapsulation sub-module receives T from the timestamp synchronization and alignment sub-module ij (t ′ ), and adopts the data compression coding rule S k =Φ(T ij (t ′ ),R) to generate a standardized data packet, where R is a preset coding resolution parameter, and Φ(·) is a compression function based on Huffman coding, and outputs a standardized data stream S k .

[0021] Preferably, in the edge computing and cloud collaborative computing module, it further includes:

[0022] The edge computing sub-module receives the standardized data stream S output by the data acquisition and compatibility management module k , and based on the structured data table T ij (t ′ ) after timestamp alignment, performs sliding window downsampling processing:

[0023]

[0024] where D m (t′) is the downsampled data block, N is the number of data points within the time, t′ n is the nth aligned time point, t′ m is the mth aligned time point, and δ(·) is the window function;

[0025] The cloud collaborative computing sub-module receives D m (t′) from the edge computing sub-module, and constructs an autoregressive prediction model based on the time series X(t)={V i (t′),I j (t′),T k (t′)}:

[0026]

[0027] where φ0 is the DC component, P is the autoregressive order, L is the sliding window length, and ∈(t) is the residual term, and generates an energy efficiency optimization strategy E(t) for grid equipment:

[0028]

[0029] where X(t - pL) is the set of voltage, current, and temperature parameters at time t - pL, and φ p is the autoregressive coefficient.

[0030] Preferably, in the intelligent fault prediction and preventive maintenance module, it further includes:

[0031] The fault prediction sub-module uses a long short-term memory network to predict the fault trend of the device. The state update formula of the long short-term memory network includes:

[0032] f t =σ(W f ·[h t-1 ,x t +b f ),

[0033] i t =σ(W i ·[h t-1 ,x t +b i ),

[0034]

[0035] o t =σ(W o ·[h t-1 ,x t +b o ),

[0036] h t =o t ⊙tanh(C t ),

[0037] where f t is the forget gate, i t is the input gate, is the candidate memory, C t is the current memory, o t is the output gate, h t is the hidden state, σ is the activation function, W f is the weight matrix of the forget gate, h t-1 is the hidden state of the previous moment, x t is the input vector of the current moment, b f is the bias term of the forget gate, W i is the weight matrix of the input gate, b i is the bias term of the input gate, tanh is the activation function, W C is the weight matrix of the candidate memory, b C is the bias term of the candidate memory, C t-1 is the state of the memory cell of the previous moment, ⊙ is the element-wise multiplication, W o is the weight matrix of the output gate, b o is the bias term of the output gate;

[0038] The preventive maintenance decision sub-module generates an optimal maintenance strategy based on the prediction results to minimize the total cost J, and the expression is: J=C m+γC f ,

[0039] where C m is the maintenance cost, C f is the failure cost, and γ is the adjustment coefficient.

[0040] Preferably, in the data security module, it further includes:

[0041] A blockchain storage sub-module for recording the transmission path of device data;

[0042] A differential privacy protection sub-module that uses the Laplace noise mechanism to perform differential privacy protection on the standardized data, and the noise distribution formula is:

[0043] where ∈ is the privacy budget, Δf is the data sensitivity, is the noise generated in the Laplace distribution.

[0044] Preferably, in the resource recovery and reuse closed-loop management module, it further includes:

[0045] A recovery evaluation sub-module that receives the device condition data X(t) = {V i (t′), I j (t′), T k (t′)} recorded in the encrypted storage path of the data security module, analyzes the inflection point t c of the remaining life curve output by the failure prediction module, and calculates the device health score H:

[0046]

[0047] where V nom is the rated voltage, I nom is the rated current, α and β are the voltage and current weight coefficients, V i (t) is the voltage, I j (t) is the current, and outputs the device recovery value V r ;

[0048] A resource reuse sub-module that calls the historical data T ij (t ′ ) of the data acquisition module, extracts the operating duration t i of the components, and combines the maintenance record M i to calculate the quality score Q i :

[0049]

[0050] where Q i is the quality score, t mis the m-th maintenance time point, μ is the maintenance effect attenuation coefficient, T k,avg is the average monitored value of temperature sensor k, T th is the designed temperature tolerance threshold, ρ and θ are the maintenance weight and temperature weight, t i is the cumulative operating time of the i-th component.

[0051] Preferably, the intelligent visualization and user interaction module includes:

[0052] A data visualization sub-module that displays the real-time status R(t) of the device through time series diagrams and heat maps, where S = {R(t)|t ∈ T},

[0053] where S represents the time series of the device status, and T is the time range;

[0054] An intelligent recommendation system sub-module that generates maintenance suggestions based on the device health score calculated in the recovery evaluation sub-module.

[0055] Preferably, the edge computing sub-module of the edge computing and cloud collaborative computing module performs real-time anomaly detection through the local outlier factor, and the calculation formula of the LOF value is:

[0056]

[0057] where LOF k (X) represents the local outlier factor of the data point, k represents the number of points in the neighborhood used to calculate the local outlier factor, and lrd(X) represents the local reachability density of the data point X.

[0058] Preferably, the data acquisition and compatibility management module uses the matrix M for standardizing the format of multi-dimensional operation data, and the conversion formula is:

[0059] X standard = M·X non-standard ,

[0060] where X standard is the standardized data, X non-standard is the non-standard data, and M is the data mapping matrix.

[0061] A product life cycle management method based on the smart grid includes the following steps:

[0062] Step 1, generate a raw data set with timestamps, align the clock offsets of multi-source sensors through the time window sliding algorithm, convert heterogeneous data into a synchronous structured data table, and encapsulate it into a standardized data stream using the Huffman coding compression rule;

[0063] Step 2: Receive the standardized data stream, perform sliding window downsampling processing at the edge node, clear the noise data and trigger threshold anomaly detection, upload the cleaned data to the cloud, construct an autoregressive prediction model in combination with the historical load curve, and generate an energy efficiency optimization strategy for grid equipment;

[0064] Step 3: Based on the energy efficiency optimization strategy, call the long short-term memory network model, analyze the temporal correlation characteristics of voltage, current, and temperature parameters, predict the inflection point of the remaining life curve of the equipment, and generate a preventive maintenance plan in combination with the inspection cycle and spare part replacement plan;

[0065] Step 4: Add device fingerprint identification to the standardized data stream, use the consortium chain architecture to bind the device fingerprint and operation data and write them into the distributed ledger, and inject Laplace noise into the spare part replacement plan during the transmission process to achieve the secure storage and cross-regional transmission of encrypted data;

[0066] Step 5: When the device health degree is lower than the preset threshold, parse the material list and working condition records in the encrypted storage path, calculate the cumulative operation duration of components and the maintenance record attenuation coefficient, and generate the priority ranking of disassembly plans and the material regeneration path planning;

[0067] Step 6: Integrate the disassembly plan, material regeneration path, and encrypted data stream, reconstruct the 3D topological map of the device and overlay the fault probability heat map, generate a maintenance work order sequence according to the health degree decay rate, energy efficiency deviation, and reuse rate, and push the priority ranking result to the interactive dashboard.

[0068] The present invention provides a product life cycle management system and method based on the smart grid. It has the following beneficial effects:

[0069] 1. By integrating a variety of sensors and edge computing technologies, the present invention realizes the real-time collection and rapid processing of multi-dimensional operation data of power equipment. Its efficient data processing ability enables the system to quickly respond to changes in equipment status, timely identify potential problems, reduce equipment failure rates, and improve the operation reliability of equipment.

[0070] 2. Using advanced machine learning algorithms for fault trend prediction, the present invention enables the system to identify the fault risks of equipment in advance. Its prediction ability can help enterprises formulate reasonable preventive maintenance plans, reduce downtime and maintenance costs caused by sudden failures, and improve the overall availability of equipment.

[0071] 3. Adopting blockchain storage and differential privacy protection mechanisms, the present invention effectively guarantees the security of device data and user privacy. The blockchain technology ensures the immutability of data, while differential privacy protection provides an additional security level for sensitive data, enabling the data to be fully protected during transmission and storage.

[0072] 4. By implementing closed-loop management of resource recovery and reuse, the system of the present invention can effectively evaluate the recovery value and reuse potential of equipment at the end of its life cycle. This process reduces resource waste, promotes sustainable development, meets modern environmental protection requirements, and helps achieve efficient utilization of resources.

[0073] 5. Through chart and real-time data display, the present invention helps users more intuitively understand the equipment status and operation conditions. At the same time, the maintenance suggestions generated by the system provide reliable decision-making support for users, making the maintenance work more scientific and systematic, and improving the operation efficiency and satisfaction of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 It is a system framework diagram of a product life cycle management system based on the smart grid according to the present invention;

[0075] Figure 2 It is a flow chart of a product life cycle management method based on the smart grid according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0076] To enable those skilled in the art to understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0077] The present invention will be described in detail below with reference to the accompanying drawings:

[0078] Embodiment:

[0079] Please refer to the attached Figure 1 , the embodiment of the present invention provides a product life cycle management system based on the smart grid. The product life cycle management system includes:

[0080] A data acquisition and compatibility management module, which collects voltage, current, and temperature data through a sensor network, converts them into a structured data table with time stamps, and outputs standardized data;

[0081] An edge computing and cloud collaborative computing module, which receives the standardized data, performs data cleaning and denoising preprocessing at the edge node, triggers threshold anomaly detection, and uploads the cleaned data to the cloud to generate an energy efficiency optimization strategy;

[0082] An intelligent fault prediction and preventive maintenance module, which constructs an equipment health model based on the generated energy efficiency optimization strategy, and outputs the remaining life curve, inspection cycle, and spare part replacement plan;

[0083] Data security module, adding device fingerprints to standardized data, binding fingerprints with data through blockchain, processing spare part replacement plans using differential privacy mechanism, and outputting encrypted data and cloud storage paths;

[0084] Closed-loop management module for resource recycling and reuse, parsing the material list and working condition records according to the remaining life curve and encrypted storage path, and generating disassembly plans and material regeneration paths when the device health threshold is triggered;

[0085] Intelligent visualization and user interaction module, integrating disassembly plans, material regeneration paths and encrypted data, constructing 3D topology maps and fault heat maps, and pushing maintenance suggestions to the interactive dashboard.

[0086] Data acquisition and compatibility management module, further including:

[0087] Multi-source heterogeneous data acquisition sub-module, collecting voltage V i (t), current I j (t), temperature T k (t) through the sensor network, and outputting the original data set D ij with timestamps:

[0088] D ij = {(V i (t), I j (t), T k (t), t)},

[0089] where i, j, k represent sensor numbers, and t is the acquisition timestamp;

[0090] Timestamp synchronization and alignment sub-module, receiving D output by the multi-source heterogeneous data acquisition sub-module ij , and aligning the multi-sensor timestamps through the time window sliding algorithm t ′ = t + τ i , where τ i is the clock offset of the i-th sensor, and t ′ is the unified timestamp after alignment, generating the synchronized structured data table T ij (t ′ );

[0091] Structured data encapsulation sub-module, receiving T ij (t ′ ) from the timestamp synchronization and alignment sub-module, and using the data compression coding rule S k = Φ(T ij (t ′ ), R) to generate a standardized data packet, where R is a preset coding resolution parameter, and Φ(·) is a compression function based on Huffman coding, outputting the standardized data stream S k .

[0092] In the edge computing and cloud collaborative computing module, it further includes:

[0093] The edge computing sub-module receives the standardized data stream S output by the data acquisition and compatibility management module k , and based on the structured data table T after timestamp alignment ij (t ′ ), performs sliding window downsampling processing:

[0094]

[0095] where D m (t′) is the downsampled data block, N is the number of data points within the time, t′ n is the nth aligned time point, t′ m is the mth aligned time point, δ(·) is the window function;

[0096] The cloud collaborative computing sub-module receives D m (t′) from the edge computing sub-module, and constructs an autoregressive prediction model based on the time series X(t) = {V i (t′), I j (t′), T k (t′)}:

[0097]

[0098] where φ0 is the DC component, P is the autoregressive order, L is the sliding window length, ∈(t) is the residual term, and generates the grid equipment energy efficiency optimization strategy E(t):

[0099]

[0100] where X(t - pL) is the set of voltage, current, and temperature parameters at time t - pL, and φ p is the autoregressive coefficient.

[0101] In the intelligent fault prediction and preventive maintenance module, it further includes:

[0102] The fault prediction sub-module uses a long short-term memory network to predict the fault trend of the equipment. Among them, the state update formula of the long short-term memory network includes:

[0103] f t = σ(W f ·[h t-1 , x t +b f ),

[0104] i t = σ(Wi ·[h t-1 ,x t +b i ),

[0105]

[0106] o t =σ(W o ·[h t-1 ,x t +b o ),

[0107] h t =o t ⊙tanh(C t ),

[0108] where f t is the forget gate, i t is the input gate, is the candidate memory, C t is the current memory, o t is the output gate, h t is the hidden state, σ is the activation function, W f is the weight matrix of the forget gate, h t-1 is the hidden state at the previous moment, x t is the input vector at the current moment, b f is the bias term of the forget gate, W i is the weight matrix of the input gate, b i is the bias term of the input gate, tanh is the activation function, W C is the weight matrix of the candidate memory, b C is the bias term of the candidate memory, C t-1 is the memory cell state at the previous moment, ⊙ is the element-wise multiplication, W o is the weight matrix of the output gate, b o is the bias term of the output gate;

[0109] The preventive maintenance decision sub-module generates the optimal maintenance strategy based on the prediction results to minimize the total cost J, and the expression is: J = C m +γC f ,

[0110] where C m is the maintenance cost, C f is the failure cost, and γ is the adjustment coefficient.

[0111] In the data security module, it further includes:

[0112] The blockchain storage sub-module is used to record the transmission path of device data;

[0113] The differential privacy protection sub-module uses the Laplace noise mechanism to perform differential privacy protection on the standardized data. The noise distribution formula is:

[0114] where ∈ is the privacy budget, Δf is the data sensitivity, is the noise generated in the Laplace distribution.

[0115] In the closed-loop management module for resource recovery and reuse, it further includes:

[0116] The recovery evaluation sub-module receives the device condition data X(t) = {V i (t′), I j (t′), T k (t′)} recorded in the encrypted storage path of the data security module, analyzes the inflection point t c of the remaining life curve output by the fault prediction module, and calculates the device health score H:

[0117]

[0118] where V nom is the rated voltage, I nom is the rated current, α and β are the voltage and current weight coefficients, V i (t) is the voltage, I j (t) is the current, and outputs the device recovery value V r ;

[0119] The resource reuse sub-module calls the historical data T ij (t ′ ) of the data acquisition module, extracts the operation duration t i of the components, and combines the maintenance record M i to calculate the quality score Q i :

[0120]

[0121] where Q i is the quality score, t m is the mth maintenance time point, μ is the maintenance effect attenuation coefficient, T k,avg is the average monitored value of the temperature sensor k, T th is the designed temperature tolerance threshold, ρ and θ are the maintenance weight and temperature weight, and t i is the cumulative operation time of the ith component.

[0122] The intelligent visualization and user interaction module includes:

[0123] Data visualization sub-module, which displays the real-time status R(t) of the device through time series diagrams and heat maps, where S = {R(t)|t ∈ T},

[0124] where S represents the time series of device status, and T is the time range;

[0125] Intelligent recommendation system sub-module, which generates maintenance suggestions based on the device health score calculated in the recovery evaluation sub-module.

[0126] The edge computing sub-module of the edge computing and cloud collaborative computing module performs real-time anomaly detection through the local outlier factor. The calculation formula for the LOF value is:

[0127]

[0128] where LOF k (X) represents the local outlier factor of the data point, k represents the number of points in the neighborhood used to calculate the local outlier factor, and lrd(X) represents the local reachability density of the data point X.

[0129] The data acquisition and compatibility management module uses the matrix M to standardize the format of multi-dimensional operation data. The conversion formula is:

[0130] X standard = M · X non-standard ,

[0131] where X standard is the standardized data, X non-standard is the non-standard data, and M is the data mapping matrix.

[0132] The data acquisition and compatibility management module effectively solves the problem of fragmented raw data caused by diverse sensor types and clock asynchronization in the smart grid through multi-source heterogeneous data acquisition and timestamp synchronization technology. The standardized data stream provides a unified input interface for subsequent analysis. The Huffman coding compression mechanism reduces the transmission bandwidth occupancy while ensuring data accuracy, significantly improving the acquisition efficiency of massive power device data and the system compatibility, and laying a high-quality data foundation for the whole life cycle management.

[0133] The edge computing and cloud collaborative computing module completes data cleaning and preliminary analysis locally through sliding window downsampling and real-time anomaly detection, significantly reducing the cloud computing load and shortening the response delay; the cloud collaborative computing sub-module uses the autoregressive model to predict the energy efficiency trend of long-period data, realizes the dynamic optimization of the power grid load and the generation of energy scheduling strategies, forming a "real-time edge processing - in-depth cloud analysis" collaborative architecture, taking into account both the computing efficiency and the global decision-making ability.

[0134] The intelligent fault prediction and preventive maintenance module, a fault prediction model based on long short-term memory network, captures the temporal correlation features of voltage, current, and temperature parameters to accurately identify the decay trend of equipment health. Combining with the dynamic trade-off model of maintenance cost and fault risk, it generates spare part replacement plans and inspection cycle suggestions, transforming traditional passive maintenance into predictive maintenance, extending the service life of equipment and reducing operation and maintenance costs by more than 30%.

[0135] The data security module. Blockchain technology ensures the full-link traceability and anti-tampering of device data from collection to storage, and the differential privacy mechanism protects the privacy and security of sensitive information through Laplace noise injection. The two are combined to build a dual protection system of "trusted data transmission - compliant privacy use" to meet the strict requirements of the power industry for data security and privacy protection.

[0136] The closed-loop management module for resource recovery and reuse. By analyzing the equipment material list and historical operating condition data, it quantitatively evaluates the remaining value and material regeneration potential of components, generates disassembly priority plans and regeneration path planning, and increases the resource recovery rate of metals, plastics, etc. of retired equipment to more than 90%, promoting the transformation of the smart grid from a linear "production - scrapping" mode to a green closed-loop mode of "resource recycling and utilization".

[0137] The intelligent visualization and user interaction module. The 3D topology map and fault heat map intuitively display the equipment operation status and risk distribution. The maintenance work order priority deduction model combines health status, energy efficiency deviation, and resource reuse rate to generate decision-making suggestions, reducing the complexity of manual judgment, enabling operation and maintenance personnel to quickly locate problems and formulate action plans, and improving management efficiency and decision-making scientificity.

[0138] Please refer to the appendix Figure 2 , a product life cycle management method based on a smart grid, including the following steps:

[0139] Step 1, generate a timestamped original data set, align the clock offsets of multi-source sensors through a time window sliding algorithm, convert heterogeneous data into a synchronized structured data table, and encapsulate it into a standardized data stream using Huffman coding compression rules;

[0140] Step 2, receive the standardized data stream, perform sliding window downsampling processing at the edge node, remove noise data and trigger threshold anomaly detection, upload the cleaned data to the cloud, and build an autoregressive prediction model in combination with the historical load curve to generate an energy efficiency optimization strategy for grid equipment;

[0141] Step 3, based on the energy efficiency optimization strategy, call the long short-term memory network model, analyze the temporal correlation features of voltage, current, and temperature parameters, predict the inflection point of the equipment remaining life curve, and generate a preventive maintenance plan in combination with the inspection cycle and spare part replacement plan;

[0142] Step 4: Add device fingerprint identification to the standardized data stream, use the consortium blockchain architecture to bind the device fingerprint to the operation data and write it into the distributed ledger, and inject Laplace noise into the spare part replacement plan during the transmission process to achieve the secure storage and cross-regional transmission of encrypted data;

[0143] Step 5: When the device health level is lower than the preset threshold, parse the material list and working condition records in the encrypted storage path, calculate the cumulative operation duration of the components and the maintenance record decay coefficient, and generate the priority ranking of the disassembly plan and the material regeneration path planning;

[0144] Step 6: Integrate the disassembly plan, material regeneration path and encrypted data stream, reconstruct the three-dimensional topological map of the device and overlay the fault probability heat map, generate a maintenance work order sequence according to the health decay rate, energy efficiency deviation and reuse rate, and push the priority ranking result to the interactive dashboard.

[0145] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principle and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A product life cycle management system based on a smart grid, characterized in that The product life cycle management system includes: A data acquisition and compatibility management module that collects voltage, current, and temperature data through a sensor network, converts it into a structured data table with timestamps, and outputs standardized data; An edge computing and cloud collaborative computing module that receives the standardized data, performs data cleaning and denoising preprocessing at the edge node, triggers threshold anomaly detection, and uploads the cleaned data to the cloud to generate an energy efficiency optimization strategy; An intelligent fault prediction and preventive maintenance module that constructs a device health model based on the generated energy efficiency optimization strategy and outputs the remaining life curve, inspection cycle, and spare part replacement plan; A data security module that adds a device fingerprint to the standardized data, binds the fingerprint to the data through blockchain, and processes the spare part replacement plan using the differential privacy mechanism to output encrypted data and the cloud storage path; A resource recovery and reuse closed-loop management module that, based on the remaining life curve and the encrypted storage path, parses the material list and operating conditions record when the device health threshold is triggered, and generates a disassembly plan and a material regeneration path; An intelligent visualization and user interaction module that integrates the disassembly plan, material regeneration path, and encrypted data, constructs a 3D topology map and a fault heat map, and pushes maintenance suggestions to the interactive dashboard.

2. The product life cycle management system based on the smart grid according to claim 1, wherein The data acquisition and compatibility management module further includes: The multi-source heterogeneous data acquisition sub-module collects the voltage V i (t), current I j (t), temperature T k (t), and outputs the original data set D with timestamps ij : D ij = {(V i (t), I j (t), T k (t), t)}, where i, j, k represent sensor numbers, and t is the acquisition timestamp; Timestamp synchronization and alignment sub-module, which receives D output by the multi-source heterogeneous data acquisition sub-module ij , and aligns the multi-sensor timestamps through the time window sliding algorithm t ′ = t + τ i , where τ i is the clock offset of the i-th sensor, and t ′ is the unified timestamp after alignment, generating a synchronized structured data table T ij (t ′ ); Structured data encapsulation sub-module, which receives T from the timestamp synchronization alignment sub-module ij (t ′ ), and uses the data compression encoding rule S k =Φ(T ij (t ′ ), R) to generate a standardized data packet, where R is a preset encoding resolution parameter, and Φ(·) is a compression function based on Huffman coding, and outputs a standardized data stream S k .

3. The product life cycle management system based on the smart grid according to claim 1, characterized in that, In the edge computing and cloud collaborative computing module, it further includes: The edge computing sub-module receives the standardized data stream S output by the data acquisition and compatibility management module k , and based on the structured data table T after timestamp alignment ij (t ′ ), performs sliding window downsampling processing: Among them, D m (t′) is the downsampled data block, N is the number of data points within the time, and t′ n is the nth alignment time point, t′ m is the mth alignment time point, and δ(·) is the window function; Cloud collaborative computing sub-module, receiving D from the edge computing sub-module m (t′), and constructing an autoregressive prediction model based on the time series X(t) = {V i (t′), I j (t′), T k (t′)}: where φ0 is the DC component, P is the autoregressive order, L is the sliding window length, ∈(t) is the residual term, and the energy efficiency optimization strategy E(t) for grid equipment is generated: Among them, X(t - pL) is the set of voltage, current, and temperature parameters at time t - pL, and φ p is the autoregressive coefficient.

4. A product life cycle management system based on a smart grid according to claim 1, characterized in that, In the intelligent fault prediction and preventive maintenance module, it further includes: A fault prediction sub-module that uses a long short-term memory network to predict the fault trend of the device. The state update formula of the long short-term memory network includes: f t = σ(W f · [h t-1 , x t + b f ), i t = σ(W i · [h t-1 , x t + b i ), o t = σ(W o · [h t-1 , x t + b o ), h t = o t ⊙tanh(C t ) Among them, f t is the forget gate, i t is the input gate, is the candidate memory, C t is the current memory, o t is the output gate, h t is the hidden state, σ is the activation function, W f is the weight matrix of the forget gate, h t-1 is the hidden state at the previous moment, x t is the input vector at the current moment, b f is the bias term of the forget gate, W i is the weight matrix of the input gate, b i is the bias term of the input gate, tanh is the activation function, W C is the weight matrix of the candidate memory, b C is the bias term of the candidate memory, C t-1 is the memory cell state at the previous moment, ⊙ is the element-wise multiplication, W o is the weight matrix of the output gate, b o is the bias term of the output gate; The preventive maintenance decision sub-module generates an optimal maintenance strategy based on the prediction results to minimize the total cost J, and the expression is: J = C m + γC f , Among them, C m is the maintenance cost, C f is the failure cost, and γ is the adjustment coefficient.

5. The product life cycle management system based on the smart grid according to claim 1, characterized in that In the data security module, it further includes: A blockchain storage sub-module for recording the transmission path of device data; Differential privacy protection sub-module, which uses the Laplace noise mechanism to perform differential privacy protection on the standardized data, and the noise distribution formula is: where ∈ is the privacy budget and Δf is the data sensitivity, is the noise generated in the Laplace distribution.

6. The product life cycle management system based on the smart grid according to claim 1, wherein In the resource recovery and reuse closed-loop management module, it further includes: Recovery evaluation sub-module, which receives the equipment operating condition data X(t) = {V i (t′), I j (t′), T k (t′)} recorded in the encrypted storage path of the data security module, parses the inflection point t c of the remaining life curve output by the fault prediction module, and calculates the equipment health score H: Among them, V nom is the rated voltage, I nom is the rated current, α and β are the voltage and current weight coefficients, V i (t) is the voltage, I j (t) is the current, and the recovery value V r of the output device; The resource reuse sub-module calls the historical data T of the data acquisition module ij (t ′ ), extracts the operation duration t of the component i , combines with the maintenance record M i to calculate the quality score Q i : Among them, Q i is the quality score, t m is the m-th maintenance time point, μ is the maintenance effect attenuation coefficient, T k,avg is the average monitored value of the temperature sensor k, T th is the designed temperature tolerance threshold, ρ and θ are the maintenance weight and the temperature weight, t i is the cumulative operating time of the i-th component.

7. A product life cycle management system based on a smart grid according to claim 1, characterized in that, The intelligent visualization and user interaction module includes: A data visualization sub-module that displays the real-time state R(t) of the device through a time series diagram and a heat map, where S = {R(t)|t ∈ T}, where S represents the time series of device states, and T is the time range; An intelligent recommendation system sub-module that generates maintenance suggestions based on the device health score calculated in the recycling evaluation sub-module.

8. The product life cycle management system based on the smart grid according to claim 3, characterized in that, The edge computing sub-module of the edge computing and cloud collaborative computing module performs real-time anomaly detection through the local outlier factor. The LOF value calculation formula is: Among them, LOF k (X) represents the local outlier factor of the data point, k represents the number of points in the neighborhood used to calculate the local outlier factor, and lrd(X) represents the local reachability density of the data point X.

9. The product life cycle management system based on the smart grid according to claim 2, characterized in that The data acquisition and compatibility management module uses a matrix M for standardizing the format of multi-dimensional operation data. The conversion formula is: X standard = M·X non-standard , Among them, X standard is standardized data, X non-standard is non-standard data, and M is a data mapping matrix.

10. A product life cycle management method based on a smart grid, according to any one of claims 1 to 9, a product life cycle management system based on a smart grid, characterized in that, It includes the following steps: Step 1, generate a raw data set with timestamps, align the clock offsets of multi-source sensors through a time window sliding algorithm, convert heterogeneous data into a synchronous structured data table, and encapsulate it into a standardized data stream using the Huffman coding compression rule; Step 2: Receive the standardized data stream, perform sliding window downsampling processing at the edge node, clear the noise data and trigger threshold anomaly detection, upload the cleaned data to the cloud, construct an autoregressive prediction model in combination with the historical load curve, and generate an energy efficiency optimization strategy for grid equipment; Step 3: Based on the energy efficiency optimization strategy, call the long short-term memory network model, analyze the temporal correlation characteristics of voltage, current, and temperature parameters, predict the inflection point of the remaining life curve of the equipment, and generate a preventive maintenance plan in combination with the inspection cycle and spare part replacement plan; Step 4: Add device fingerprint identification to the standardized data stream, use the consortium chain architecture to bind the device fingerprint and operation data and write them into the distributed ledger, inject Laplace noise into the spare part replacement plan during transmission, and achieve secure storage and cross-regional transmission of encrypted data; Step 5: When the device health level is lower than the preset threshold, parse the material list and working condition records in the encrypted storage path, calculate the cumulative operation duration of components and the maintenance record decay coefficient, and generate the disassembly plan priority ranking and material regeneration path planning; Step 6: Integrate the disassembly plan, material regeneration path, and encrypted data stream, reconstruct the 3D topological map of the device and overlay the fault probability heat map, generate a maintenance work order sequence according to the health decay rate, energy efficiency deviation, and reuse rate, and push the priority ranking result to the interactive dashboard.

Images