CN117994076A - Natural gas pipe network scheduling energy-saving assessment method based on big data - Google Patents
Natural gas pipe network scheduling energy-saving assessment method based on big data Download PDFInfo
- Publication number
- CN117994076A CN117994076A CN202410208326.7A CN202410208326A CN117994076A CN 117994076 A CN117994076 A CN 117994076A CN 202410208326 A CN202410208326 A CN 202410208326A CN 117994076 A CN117994076 A CN 117994076A
- Authority
- CN
- China
- Prior art keywords
- data
- pipe network
- natural gas
- energy
- gas pipe
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 title claims abstract description 98
- 238000000034 method Methods 0.000 title claims abstract description 49
- 239000003345 natural gas Substances 0.000 title claims abstract description 49
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 57
- 238000005457 optimization Methods 0.000 claims abstract description 34
- 238000005265 energy consumption Methods 0.000 claims abstract description 23
- 238000012545 processing Methods 0.000 claims abstract description 13
- 238000012502 risk assessment Methods 0.000 claims abstract description 13
- 238000010801 machine learning Methods 0.000 claims abstract description 11
- 238000012544 monitoring process Methods 0.000 claims abstract description 10
- 230000008713 feedback mechanism Effects 0.000 claims abstract description 5
- 238000004458 analytical method Methods 0.000 claims description 28
- 238000007405 data analysis Methods 0.000 claims description 17
- 238000003066 decision tree Methods 0.000 claims description 15
- 238000007726 management method Methods 0.000 claims description 14
- 238000013135 deep learning Methods 0.000 claims description 13
- 230000000694 effects Effects 0.000 claims description 12
- 238000009826 distribution Methods 0.000 claims description 11
- 238000004134 energy conservation Methods 0.000 claims description 11
- 238000013528 artificial neural network Methods 0.000 claims description 8
- 238000005516 engineering process Methods 0.000 claims description 7
- 230000007774 longterm Effects 0.000 claims description 7
- 238000000137 annealing Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000004140 cleaning Methods 0.000 claims description 6
- 230000002159 abnormal effect Effects 0.000 claims description 5
- 238000009499 grossing Methods 0.000 claims description 5
- 238000012795 verification Methods 0.000 claims description 5
- 238000011156 evaluation Methods 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000012300 Sequence Analysis Methods 0.000 claims description 3
- 238000007418 data mining Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000000513 principal component analysis Methods 0.000 claims description 3
- 238000004088 simulation Methods 0.000 claims description 3
- 230000005540 biological transmission Effects 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 230000004044 response Effects 0.000 description 5
- 238000012360 testing method Methods 0.000 description 5
- 238000013136 deep learning model Methods 0.000 description 4
- 230000007613 environmental effect Effects 0.000 description 4
- 238000005259 measurement Methods 0.000 description 4
- 230000007246 mechanism Effects 0.000 description 4
- 230000000737 periodic effect Effects 0.000 description 4
- 239000007789 gas Substances 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 239000002699 waste material Substances 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000013486 operation strategy Methods 0.000 description 2
- 238000013439 planning Methods 0.000 description 2
- 230000003449 preventive effect Effects 0.000 description 2
- 230000001932 seasonal effect Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 239000008186 active pharmaceutical agent Substances 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- QVFWZNCVPCJQOP-UHFFFAOYSA-N chloralodol Chemical compound CC(O)(C)CC(C)OC(O)C(Cl)(Cl)Cl QVFWZNCVPCJQOP-UHFFFAOYSA-N 0.000 description 1
- 230000003749 cleanliness Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 230000003750 conditioning effect Effects 0.000 description 1
- 238000007727 cost benefit analysis Methods 0.000 description 1
- 238000012043 cost effectiveness analysis Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 230000005641 tunneling Effects 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of natural gas scheduling, in particular to a natural gas pipe network scheduling energy-saving assessment method based on big data. The method comprises the steps of monitoring the state of the whole pipe network by utilizing a sensor network, and collecting the data of the natural gas pipe network; analyzing the collected data using big data techniques and machine learning algorithms; based on the history and real-time data, performing dynamic risk assessment by using a prediction model, and predicting and preventing risk events; according to the model prediction result, an optimal energy-saving scheme is found out through an optimization algorithm; deploying edge computing equipment to realize near-real-time data processing and decision making; establishing a feedback mechanism, and continuously optimizing a prediction model and a scheduling strategy; and generating a comprehensive report regularly. The invention not only can effectively reduce the energy consumption of the natural gas pipe network and improve the energy utilization efficiency, but also can enhance the stability and the safety of the system, thereby providing powerful technical support and decision basis for the operation management of the natural gas pipe network.
Description
Technical Field
The invention relates to the technical field of natural gas scheduling, in particular to a natural gas pipe network scheduling energy-saving assessment method based on big data.
Background
In the energy field, natural gas has become an important component of global energy consumption due to its cleanliness, efficiency and economy. With the continuous increase of natural gas consumption, higher requirements are put on the dispatching and management of natural gas pipe networks. Traditional natural gas pipeline network scheduling focuses on meeting gas requirements and guaranteeing safe operation of the pipeline network, and is relatively less in energy-saving efficiency. Currently, the following problems also exist: for maintenance and safety of a pipe network, the traditional method generally relies on periodic inspection and experience feedback of past accidents, and the method cannot monitor the condition of the pipe network in real time, so that delay discovery problems can be caused; the prior art fails to fully utilize data analysis and machine learning algorithms to identify energy saving potential and optimize operation strategies, resulting in energy waste and higher operation cost; the traditional method lacks an effective dynamic risk assessment mechanism and an optimization mechanism, is difficult to predict and prevent risk events in time, and influences the reliability and safety of a natural gas pipe network.
Disclosure of Invention
In order to solve the problems, the invention provides a natural gas pipe network scheduling energy-saving assessment method based on big data.
In order to achieve the above purpose, the invention adopts the following technical scheme:
a natural gas pipe network scheduling energy-saving assessment method based on big data comprises the following steps:
monitoring the state of the whole pipe network by using a sensor network, and collecting the data of the natural gas pipe network;
analyzing the collected data using big data techniques and machine learning algorithms;
Based on the history and real-time data, performing dynamic risk assessment by using a prediction model, and predicting and preventing risk events;
according to the model prediction result, an optimal energy-saving scheme is found out through an optimization algorithm;
deploying edge computing equipment to realize near-real-time data processing and decision making;
establishing a feedback mechanism, and continuously optimizing a prediction model and a scheduling strategy;
And (3) generating a comprehensive report regularly, and making a long-term energy management and investment strategy to provide decision support.
Further, the data includes pipe network operation data and environmental data.
Further, the analysis of the collected data using big data techniques and machine learning algorithms includes the steps of:
acquiring collected data, and performing data cleaning, interpolation and data standardization or normalization operation on the data;
applying an autoregressive model to perform time sequence analysis;
Extracting periodicity by using Fourier transform, and reducing data dimension by using principal component analysis;
and (3) carrying out data analysis by using a decision tree and a deep learning algorithm, optimizing by adopting a Bayesian algorithm, and generating an analysis result.
Furthermore, the data analysis is performed by using a decision tree and a deep learning algorithm, the optimization is performed by adopting a Bayesian algorithm, and an analysis result is generated, and the method comprises the following steps:
carrying out preliminary classification and analysis on the collected data by adopting a decision tree algorithm, and identifying key features and potential modes in the data;
further analyzing the key features and modes by using a deep learning algorithm, and extracting complex data relationship and deep information by using a multi-layer neural network;
modeling and optimizing by using a Bayesian algorithm, and improving the prediction capability of the model by using priori knowledge and posterior probability;
Comprehensively utilizing analysis results of decision trees, deep learning and Bayesian algorithm to form a comprehensive data analysis model;
And generating analysis results and suggestions through the comprehensive data analysis model.
Further, the dynamic risk assessment is performed by using a prediction model based on historical and real-time data, and the risk event is predicted and prevented, and the method comprises the following steps:
integrating and preprocessing data;
extracting key features from the data by using a data mining technology;
dynamic weight distribution is carried out on the key features by using a weighted moving average or an exponential smoothing method;
Identifying an abnormal pattern in the data using the self-encoder;
Analyzing risk factors by using a gradient elevator and a neural network algorithm;
and quantifying the predicted risk, generating a prediction result, and classifying according to the risk level.
Further, the risk level sequentially comprises a high risk, a medium risk and a low risk, and the risk levels are classified according to risk indexes, wherein the risk indexes comprise a pressure index, a flow index, a temperature index and an equipment failure index.
Further, the method for finding out the optimal energy-saving scheme through the optimization algorithm according to the model prediction result comprises the following steps:
constructing an energy consumption model, and taking energy consumption data of a natural gas pipe network and related operation parameters as input variables;
designing a quantum optimization algorithm, adopting quantum bits to represent decision variables in a scheduling scheme, and searching an optimal pipe network scheduling strategy by utilizing superposition and entanglement characteristics of quantum states;
Carrying out quantum annealing or quantum approximation optimization algorithm, carrying out iterative optimization on the energy consumption model and the scheduling strategy, and quickly approaching to a global optimal solution through probability amplitude adjustment of quantum calculation;
The result analysis and verification are carried out by combining a classical computer, and the energy-saving scheme obtained by a quantum optimization algorithm is decoded and the efficiency is evaluated by utilizing a quantum-classical mixed computing framework;
According to the quantum calculation optimization result, the pipe network operation mode and parameter setting are dynamically adjusted, energy-saving measures are implemented, and the energy-saving effect is verified through simulation or actual application.
Further, the edge computing device adopts a processor based on ARM architecture and a real-time operating system, and combines a data bus and a network protocol.
Further, the integrated report includes energy saving effects, operational stability and safety assessment.
The invention has the beneficial effects that:
According to the invention, the optimal energy-saving scheme is found out by using an optimization algorithm, so that the energy consumption in the natural gas conveying process can be effectively reduced, unnecessary energy waste is reduced, and the overall energy efficiency is improved. Through using big data technology and machine learning algorithm to carry out deep analysis to the operation data of natural gas pipe network, can discern the potential opportunity that the energy efficiency promoted, through implementing best energy-conserving scheme, reduce the energy waste, reduce the operation cost of natural gas pipe network. The prediction model is utilized to carry out dynamic risk assessment, and possible risk events such as leakage, pressure abnormality and the like can be predicted and prevented, so that preventive measures are taken in advance, and the system reliability and safety of the natural gas pipe network are enhanced. By deploying the edge computing equipment, near-real-time data processing and decision making are realized, so that the dispatching of the natural gas pipe network is more intelligent and automatic, and the dispatching efficiency and response speed are improved. The comprehensive report generated regularly provides deep analysis on the running efficiency, risk management and energy saving effect of the natural gas pipeline network, provides powerful decision support for long-term energy management and investment strategies, and helps enterprises optimize resource allocation and future planning. By optimizing the dispatching and operation of the natural gas pipe network, the energy consumption is reduced, the environmental pressure is lightened, the enterprise is helped to achieve the environmental protection aim, and the sustainable use of energy is promoted.
Drawings
FIG. 1 is a schematic flow chart of a natural gas pipe network scheduling energy-saving assessment method based on big data in the invention.
Fig. 2 is a flowchart of step S24 according to an embodiment of the invention.
Detailed Description
Referring to fig. 1-2, the invention relates to a natural gas pipe network dispatching energy-saving evaluation method based on big data.
Examples
A natural gas pipe network scheduling energy-saving assessment method based on big data comprises the following steps:
S1: monitoring the state of the whole pipe network by using a sensor network, and collecting the data of the natural gas pipe network; the data includes pipe network operation data and environmental data.
Specifically, a temperature sensor with high precision and wide temperature range is selected and used for monitoring the temperature change of the gas in the pipeline; the pressure sensor with high stability and high response speed is adopted to monitor the air pressure state in the pipeline in real time; selecting differential pressure type, turbine type or ultrasonic flow sensors to accurately measure the flow speed and total flow of natural gas; and an online gas analyzer is deployed for monitoring the natural gas composition and ensuring the natural gas quality. And deploying key sensors at the inlet, the outlet, the important nodes and the conversion stations of the pipe network so as to comprehensively monitor the state of the pipe network. And setting reasonable data acquisition frequency according to the dynamic performance of pipe network operation and the real-time monitoring requirement. For critical areas and high risk nodes, the acquisition frequency may be higher than for normal areas. And the self-adaptive acquisition frequency technology is adopted, the acquisition frequency is automatically adjusted according to the real-time data change, and the data collection efficiency is optimized. A Low Power Wide Area Network (LPWAN) technology, such as LoRa or NB-IoT, is used to achieve long-range, low power data transmission. And an efficient network topological structure, such as a star-shaped or mesh-shaped structure, is designed to ensure the reliability and stability of data transmission. And a data encryption and security authentication mechanism is implemented, so that the data security in the transmission process is ensured. And deploying edge computing equipment at the key nodes, performing primary processing on the acquired original data, such as data cleaning, compression and feature extraction, and relieving the burden of a central server. And the edge equipment is utilized to realize local analysis and processing of data, and the emergency of the pipe network is responded quickly. And a distributed data storage system is constructed, so that high availability and fault tolerance of data are ensured. And a time sequence database is adopted to manage pipe network data, so that efficient data query and analysis are supported. And (3) implementing data life cycle management, and cleaning and archiving old data periodically according to the value and application requirements of the data. The pipe network operation data comprise parameters such as pressure, flow, temperature and the like, and the environment data comprise information such as meteorological conditions, equipment states and the like.
S2: analyzing the collected data using big data techniques and machine learning algorithms;
The step S2 specifically includes the following steps:
s21: acquiring collected data, and performing data cleaning, interpolation and data standardization or normalization operation on the data;
s22: applying an autoregressive model to perform time sequence analysis;
specifically, a smoothness test (such as ADF test) of the data is performed to ensure that the time series data satisfies the preconditions of the application of the autoregressive model. And selecting optimal model parameters by iteratively testing different parameter combinations by using the standards such as the red pool information criterion (AIC) or the Bayesian Information Criterion (BIC). The time series data are divided into a training set and a testing set, the training set is firstly used for estimating model parameters, and then the prediction performance of the model is verified on the testing set so as to evaluate the accuracy and generalization capability of the model. The model parameters are re-estimated using the selected optimal model and the ensemble of data, and predictions of future points in time are made. And analyzing the prediction result, identifying possible risk points or optimization opportunities, and providing decision support for pipe network operation.
S23: extracting periodicity by using Fourier transform, and reducing data dimension by using principal component analysis;
Specifically, time-series data with suspected periodic characteristics, such as daily consumption, seasonal variation, and the like, are selected. Fourier transforming the selected time series data is performed to transform the time series from the time domain to the frequency domain to identify the dominant frequency components in the data. Analysis of the transformation results identifies and extracts significant periodic frequency components that characterize the primary periodic variation of the data.
S24: and (3) carrying out data analysis by using a decision tree and a deep learning algorithm, optimizing by adopting a Bayesian algorithm, and generating an analysis result.
Further, the step S24 specifically includes the following steps:
s241: carrying out preliminary classification and analysis on the collected data by adopting a decision tree algorithm, and identifying key features and potential modes in the data;
It should be noted that a decision tree algorithm (e.g., CART or C4.5) is selected to construct a classification or regression model. Cross-validation methods are used to evaluate the generalization ability of the model and to select the best parameter settings (e.g., tree depth, minimum number of split samples, etc.).
S242: further analyzing the key features and modes by using a deep learning algorithm, and extracting complex data relationship and deep information by using a multi-layer neural network;
specifically, based on the identified key features, a deep learning model architecture, such as Convolutional Neural Network (CNN) is designed for analysis of feature imaging data, or Recurrent Neural Network (RNN) is designed for time series data analysis. Deep learning models are trained using large-scale data sets. The verification set is used to adjust the network parameters and the layer number to achieve the best performance. And extracting complex relations and deep information in the data through the trained multi-layer neural network, and further understanding the dynamic nature and complexity of pipe network operation.
S243: modeling and optimizing by using a Bayesian algorithm, and improving the prediction capability of the model by using priori knowledge and posterior probability;
Specifically, a priori knowledge about natural gas network energy savings is collected, including information obtained from historical data, expert experience, and literature. For example, based on past experience, fine-tuning of certain operating parameters may result in significant changes in energy consumption, which information is formalized as a priori distribution of the model. A probabilistic model is built to describe a relationship between observed data and unknown parameters. For example, it may be assumed that a probability distribution is followed between the energy saving effect and a specific operating parameter, the parameters of which are estimated from the data. And (3) calculating posterior distribution of the unknown parameters by applying Bayesian theorem and combining the observed data and the priori knowledge. This process typically involves complex integration operations, and numerical methods, such as Markov Chain Monte Carlo (MCMC) methods, may be employed to approximate the posterior distribution. By analyzing the posterior distribution, the most probable value (posterior mean) or other characteristic (e.g., median, confidence interval, etc.) of the unknown parameter can be obtained. On the basis, the model parameters are adjusted, and the prediction accuracy of the energy-saving strategy is optimized. And forecasting and deciding by using posterior distribution. For example, the probability of achieving a particular energy saving objective under a given operating condition may be calculated to guide the actual operation.
S244: comprehensively utilizing analysis results of decision trees, deep learning and Bayesian algorithm to form a comprehensive data analysis model, and identifying opportunities for energy conservation optimization;
Specifically, key feature recognition of the decision tree model, complex relation extraction of the deep learning model and optimization decision results of the Bayesian model are integrated. This may be achieved by model integration techniques, such as using weighted averaging, stacking, or model fusion techniques, to determine the weights of the individual models based on their performance on the validation set. Based on the fusion result, the model parameters and policies are further adjusted. For example, if the Bayesian model shows a stronger relationship of a feature identified by a decision tree to energy savings effects than expected, the weights of the feature in the deep learning model can be adjusted accordingly. And analyzing the output of the comprehensive model in detail, and identifying the factors with the most influence and the most effective energy-saving measures. For example, the model may indicate that adjusting certain parameters under certain operating conditions can significantly reduce energy consumption. And (3) formulating specific energy-saving optimization suggestions according to the analysis result of the comprehensive model. These suggestions will be based on the most efficient energy-saving measures of model identification and consider feasibility of implementation and cost-effectiveness ratio.
S245: and generating analysis results and suggestions through the comprehensive data analysis model.
S3: based on the history and real-time data, performing dynamic risk assessment by using a prediction model, and predicting and preventing risk events;
The step S2 specifically includes the following steps:
s31: integrating and preprocessing data;
s32: extracting key features from the data by using a data mining technology;
S33: dynamic weight distribution is carried out on the key features by using a weighted moving average or an exponential smoothing method;
it should be noted that the weighted moving average is applicable to when the data shows a relatively stable trend. The appropriate time window size is determined and data points within each window are assigned different weights, typically with the nearest data point having a higher weight. The exponential smoothing method is applicable to the existence of strong fluctuations or seasonal variations in the data. A smoothing parameter is set that determines the degree of importance to the most recent observations.
And (3) performing parameter tuning by using the historical data set, and finding the optimal time window and the smooth parameters so as to ensure that the prediction is as accurate as possible. The optimal parameters are determined by comparing the prediction results at different parameter settings, e.g. using the mean square error MSE. The selected method is applied to the historical data for weight distribution to calculate a weighted average or smoothed value for each point in time. The weight assignments are updated to reflect the impact of the latest data on future trends.
S34: identifying an abnormal pattern in the data using the self-encoder;
S35: analyzing risk factors by using a gradient elevator and a neural network algorithm;
Specifically, gradient hoist model parameters such as tree depth, learning rate, number of trees, regularization parameters, etc. are set. These parameters are optimized using a grid search or a random search, etc. The features are suitably processed, including encoding of the classification variables (e.g., one-hot encoding or tag encoding) and scaling of the numerical features. Processing missing values the gradient hoist model can process missing values but requires determining the optimal processing strategy. New features are created, such as interactive features, aggregated features, or statistical features based on existing data. Feature selection is performed to reject non-important features and reduce model complexity.
S36: quantifying the predicted risk, generating a prediction result, and classifying according to the risk level; the risk level sequentially comprises high risk, medium risk and low risk, and is classified according to risk indexes, wherein the risk indexes comprise a pressure index, a flow index, a temperature index and an equipment failure index.
Specifically, a gradient hoist model is used to output the risk probability for each predicted event. Each predicted event is assigned a risk score based on the risk probability. By mapping the probability to a predetermined scoring range. Different risk classes, such as high risk, medium risk and low risk, are defined according to business requirements and risk tolerance. A threshold is set for each risk level. For example, a high risk may correspond to a risk score of over 80, a medium risk of 40 to 80, and a low risk of less than 40. Specific risk indicators, such as pressure index, flow index, temperature index, and equipment failure index, are defined. These indicators reflect risk factors in different ways. Weights are assigned to the different risk indicators, reflecting the importance of each indicator in the overall risk assessment. The different risk indicators are integrated into a single risk assessment. This may involve a weighted average or a more complex mathematical model. Risk indicators and thresholds are periodically reviewed and adjusted to ensure accuracy and timeliness of risk assessment. The risk assessment results are visualized by using charts and dashboards, so that a management layer and operators can conveniently understand and monitor risk states. For high-risk events, an automatic alarm system is established, and timely notification of relevant personnel for countermeasures is ensured.
S4: according to the model prediction result, an optimal energy-saving scheme is found out through an optimization algorithm;
The step S4 specifically includes the following steps:
s41: constructing an energy consumption model, and taking energy consumption data of a natural gas pipe network and related operation parameters as input variables;
Specifically, historical energy consumption data of the natural gas pipe network are collected, wherein the historical energy consumption data comprise parameters such as temperature, pressure and flow. And cleaning and normalizing the data to ensure the quality of the input variable. Based on the collected data, an energy consumption model is established using statistical or machine learning methods. The model takes the operation parameters (such as pressure, flow rate and the like) of the pipe network as input, and the energy consumption as output, and reflects the influence of different operation parameters on the energy consumption.
S42: designing a quantum optimization algorithm, adopting quantum bits to represent decision variables in a scheduling scheme, and searching an optimal pipe network scheduling strategy by utilizing superposition and entanglement characteristics of quantum states;
S43: carrying out quantum annealing or quantum approximation optimization algorithm, carrying out iterative optimization on the energy consumption model and the scheduling strategy, and quickly approaching to a global optimal solution through probability amplitude adjustment of quantum calculation;
It should be noted that, in the quantum annealing process, the energy landscape of the system is constructed according to the energy consumption model, and each possible scheduling scheme corresponds to an energy level, so as to find the state with the lowest energy, i.e. the optimal scheduling scheme. Quantum annealing gradually transfers the quantum state from a high energy state to a low energy state by slowly lowering the effective temperature of the system. In the process, the system explores the energy landscape through quantum tunneling effect, and avoids sinking into local minimum values.
Quantum approximation optimization algorithms construct quantum circuits using a series of parameterized quantum gates (turngates), which are then tuned by classical optimization algorithms to find the best solution. In the quantum approximation optimization algorithm, the depth and parameters of the quantum circuit are optimized through a classical-quantum iterative process. The measurement results of each quantum state are used to adjust the parameters in an effort to more closely approximate the globally optimal solution in the next iteration.
And after the quantum annealing or quantum approximation optimization algorithm is finished, the probability distribution of the scheduling scheme is obtained by measuring the quantum state for a plurality of times. These measurements reflect the probability that various scheduling schemes will be optimal solutions.
And selecting a scheduling scheme with the highest occurrence probability as a final energy-saving scheme according to the measurement result. In practice, multiple high probability schemes may be selected for further classical verification and analysis.
S44: the result analysis and verification are carried out by combining a classical computer, and the energy-saving scheme obtained by a quantum optimization algorithm is decoded and the efficiency is evaluated by utilizing a quantum-classical mixed computing framework;
S45: according to the quantum calculation optimization result, the pipe network operation mode and parameter setting are dynamically adjusted, energy-saving measures are implemented, and the energy-saving effect is verified through simulation or actual application.
S5: deploying edge computing equipment to realize near-real-time data processing and decision making; the edge computing device adopts a processor based on ARM architecture and a real-time operating system, and combines a data bus and a network protocol.
Specifically, high performance processors based on the ARM architecture are chosen that are suitable for use in edge computing scenarios due to their low power consumption and high performance. These edge computing devices are deployed at critical locations of the natural gas network, such as valve stations, measurement stations, and conditioning stations. These devices are close to the data source, which may reduce delays in data transmission. Running on the edge computing device are real-time operating systems (RTOSs) that can provide fast response and efficient data processing. Real-time operating systems can ensure real-time performance of data processing and decision making, and particularly in emergency situations, fast response is of paramount importance. And data preprocessing is performed by using edge computing equipment, such as data filtering, denoising and local aggregation, so that the data volume required to be sent to a cloud or a central data center is reduced. Basic data analysis, such as monitoring the pipe network state in real time, and rapidly detecting and responding to abnormal conditions, is implemented. The edge computing devices communicate with the central control center and other devices via data buses and network protocols. Common protocols include lightweight communication protocols such as MQTT, coAP, etc. The network architecture is designed to ensure the security and reliability of data transmission, including the use of encryption techniques and security authentication mechanisms. The edge computing device uses APIs and standardized interfaces to enable the edge device to be compatible with different systems and applications. The edge computing device is regularly maintained and updated to ensure its stable operation and accuracy of data processing.
S6: establishing a feedback mechanism, and continuously optimizing a prediction model and a scheduling strategy;
s7: generating a comprehensive report regularly, and making a long-term energy management and investment strategy to provide decision support; the integrated report includes energy saving effects, operational stability and safety assessment.
Specifically, a feedback mechanism is established to compare the actual operational data with the results of the predictive model and the scheduling policy. And adjusting and optimizing the prediction model and the scheduling strategy according to the deviation of the actual operation data and the prediction result. And the prediction model is automatically adjusted and optimized by using a machine learning algorithm so as to improve the accuracy and efficiency of the prediction model. Artificial intelligence techniques, such as reinforcement learning, are applied to simulate different operating scenarios to find the optimal scheduling strategy. Performance of the scheduling policy is periodically assessed, including energy saving effects, cost benefit analysis, and risk management. And adjusting a scheduling strategy according to the evaluation result so as to continuously improve the operation efficiency and the safety of the pipe network. Comprehensive reports containing key performance indicators, operational stability and safety assessments are generated periodically. The report should include data analysis results, trend predictions, risk assessment, and policy suggestions. And utilizing the collected data and analysis results to formulate a long-term energy management and investment strategy. Advice for energy efficiency improvement, equipment upgrade plans, and possibly cost effectiveness analysis should be included in the report. Periodically communicate with related personnel to share reports and policies. Feedback and advice are encouraged to better adjust management and operational policies.
In the embodiment, through fine monitoring and data analysis, energy consumption hot spots and unreasonable places in the operation of the natural gas pipeline network can be identified, so that targeted energy-saving improvement measures are provided, and the energy utilization efficiency is remarkably improved. By analyzing the operation data of the pipe network by using a machine learning and deep learning algorithm, the deep relation between the operation parameters and the energy consumption can be found, scientific basis is provided for the pipe network scheduling, and the optimization of the operation strategy is realized. Through real-time monitoring and risk assessment, abnormal states and potential risks in the operation of a pipe network can be found in time, and preventive measures are taken, so that the stability and safety of the system are enhanced. The data is primarily processed through the edge computing equipment, so that the burden of a central server is reduced, and the data transmission and storage cost is reduced. Meanwhile, the operation and maintenance cost of the natural gas pipe network can be reduced by the optimized pipe network scheduling strategy and energy-saving measures. And comprehensively utilizing analysis results of decision trees, deep learning and Bayesian algorithm to form a high-accuracy data analysis model, and providing powerful decision support for pipe network operation and long-term energy management. By dynamically adjusting the data acquisition frequency and implementing real-time data processing, the system can quickly respond to changes and emergency events in the operation of the pipe network, and the adaptability and response capability of the pipe network are improved. The comprehensive report generated regularly not only comprises the energy saving effect, the running stability and the safety evaluation, but also can provide long-term energy management and investment strategy suggestions to help the management layer to carry out scientific decision and planning.
The above embodiments are merely illustrative of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention, and various modifications and improvements made by those skilled in the art to the technical solution of the present invention should fall within the scope of protection defined by the claims of the present invention without departing from the spirit of the design of the present invention.
Claims (9)
1. The natural gas pipe network scheduling energy-saving assessment method based on big data is characterized by comprising the following steps of:
monitoring the state of the whole pipe network by using a sensor network, and collecting the data of the natural gas pipe network;
analyzing the collected data using big data techniques and machine learning algorithms;
Based on the history and real-time data, performing dynamic risk assessment by using a prediction model, and predicting and preventing risk events;
according to the model prediction result, an optimal energy-saving scheme is found out through an optimization algorithm;
deploying edge computing equipment to realize near-real-time data processing and decision making;
establishing a feedback mechanism, and continuously optimizing a prediction model and a scheduling strategy;
And (3) generating a comprehensive report regularly, and making a long-term energy management and investment strategy to provide decision support.
2. The method for evaluating the energy conservation of natural gas pipe network scheduling based on big data according to claim 1, wherein the data comprises pipe network operation data and environment data.
3. The method for evaluating the energy conservation of the natural gas pipe network scheduling based on big data according to claim 1, wherein the analyzing the collected data by using big data technology and machine learning algorithm comprises the following steps:
acquiring collected data, and performing data cleaning, interpolation and data standardization or normalization operation on the data;
applying an autoregressive model to perform time sequence analysis;
Extracting periodicity by using Fourier transform, and reducing data dimension by using principal component analysis;
and (3) carrying out data analysis by using a decision tree and a deep learning algorithm, optimizing by adopting a Bayesian algorithm, and generating an analysis result.
4. The method for evaluating the dispatching energy conservation of the natural gas pipe network based on big data according to claim 3, wherein the data analysis is performed by using a decision tree and a deep learning algorithm, the optimization is performed by adopting a Bayesian algorithm, and an analysis result is generated, and the method comprises the following steps:
carrying out preliminary classification and analysis on the collected data by adopting a decision tree algorithm, and identifying key features and potential modes in the data;
further analyzing the key features and modes by using a deep learning algorithm, and extracting complex data relationship and deep information by using a multi-layer neural network;
modeling and optimizing by using a Bayesian algorithm, and improving the prediction capability of the model by using priori knowledge and posterior probability;
Comprehensively utilizing analysis results of decision trees, deep learning and Bayesian algorithm to form a comprehensive data analysis model;
And generating analysis results and suggestions through the comprehensive data analysis model.
5. The method for evaluating the dispatching energy conservation of the natural gas pipe network based on big data according to claim 1, wherein the method for evaluating the dynamic risk, predicting and preventing risk events based on historical and real-time data by using a prediction model comprises the following steps:
integrating and preprocessing data;
extracting key features from the data by using a data mining technology;
dynamic weight distribution is carried out on the key features by using a weighted moving average or an exponential smoothing method;
Identifying an abnormal pattern in the data using the self-encoder;
Analyzing risk factors by using a gradient elevator and a neural network algorithm;
and quantifying the predicted risk, generating a prediction result, and classifying according to the risk level.
6. The method for evaluating the energy conservation of the natural gas pipe network scheduling based on big data according to claim 5, wherein the risk level sequentially comprises high risk, medium risk and low risk, and is specifically classified according to risk indexes, wherein the risk indexes comprise a pressure index, a flow index, a temperature index and an equipment failure index.
7. The method for evaluating the energy conservation of the natural gas pipe network scheduling based on big data according to claim 1, wherein the method for finding the optimal energy conservation scheme through an optimization algorithm according to the model prediction result comprises the following steps:
constructing an energy consumption model, and taking energy consumption data of a natural gas pipe network and related operation parameters as input variables;
designing a quantum optimization algorithm, adopting quantum bits to represent decision variables in a scheduling scheme, and searching an optimal pipe network scheduling strategy by utilizing superposition and entanglement characteristics of quantum states;
Carrying out quantum annealing or quantum approximation optimization algorithm, carrying out iterative optimization on the energy consumption model and the scheduling strategy, and quickly approaching to a global optimal solution through probability amplitude adjustment of quantum calculation;
The result analysis and verification are carried out by combining a classical computer, and the energy-saving scheme obtained by a quantum optimization algorithm is decoded and the efficiency is evaluated by utilizing a quantum-classical mixed computing framework;
According to the quantum calculation optimization result, the pipe network operation mode and parameter setting are dynamically adjusted, energy-saving measures are implemented, and the energy-saving effect is verified through simulation or actual application.
8. The method for evaluating the dispatching energy conservation of the natural gas pipe network based on big data according to claim 1, wherein the edge computing equipment adopts a processor based on ARM architecture and a real-time operating system and combines a data bus and a network protocol.
9. The method for evaluating the energy conservation of the natural gas pipe network scheduling based on big data according to claim 1, wherein the comprehensive report comprises an energy conservation effect, operation stability and safety evaluation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410208326.7A CN117994076A (en) | 2024-02-26 | 2024-02-26 | Natural gas pipe network scheduling energy-saving assessment method based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410208326.7A CN117994076A (en) | 2024-02-26 | 2024-02-26 | Natural gas pipe network scheduling energy-saving assessment method based on big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117994076A true CN117994076A (en) | 2024-05-07 |
Family
ID=90897206
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410208326.7A Pending CN117994076A (en) | 2024-02-26 | 2024-02-26 | Natural gas pipe network scheduling energy-saving assessment method based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117994076A (en) |
-
2024
- 2024-02-26 CN CN202410208326.7A patent/CN117994076A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109872003B (en) | Object state prediction method, object state prediction system, computer device, and storage medium | |
| CN115578015B (en) | Sewage treatment whole process supervision method, system and storage medium based on Internet of things | |
| CN105117602B (en) | A kind of metering device running status method for early warning | |
| US20230419425A1 (en) | Method for monitoring operation of liquefied natural gas (lng) storage and internet of things system (iot) thereof | |
| CN105512448A (en) | Power distribution network health index assessment method | |
| CN109583520B (en) | State evaluation method of cloud model and genetic algorithm optimization support vector machine | |
| WO2023142424A1 (en) | Power financial service risk control method and system based on gru-lstm neural network | |
| CN109492790A (en) | Wind turbines health control method based on neural network and data mining | |
| CN117196159A (en) | Intelligent water service partition metering system based on Internet big data analysis | |
| Si et al. | Fault prediction model based on evidential reasoning approach | |
| CN117150934B (en) | Comprehensive monitoring system for transformer bushing state and online data processing method thereof | |
| CN114519923A (en) | Intelligent diagnosis and early warning method and system for power plant | |
| CN117689373A (en) | Maintenance decision support method for energy router of flexible direct-current traction power supply system | |
| CN117408162A (en) | Power grid fault control method based on digital twin | |
| JP7062505B2 (en) | Equipment management support system | |
| CN115936663A (en) | Maintenance method and device for power system | |
| CN117994076A (en) | Natural gas pipe network scheduling energy-saving assessment method based on big data | |
| Sun et al. | A data-driven framework for tunnel infrastructure maintenance | |
| Chen et al. | BIM-and IoT-Based Data-Driven Decision Support System for Predictive Maintenance of Building Facilities | |
| Sverdlova et al. | Predicting anomaly conditions of energy equipment using neural networks | |
| Shcherbatov | Current state of predictive analytics systems development in the energy sector | |
| CN113269435B (en) | New energy station running state coupling monitoring and evaluating system | |
| Ding et al. | Time Series Data Cleaning Method Based on Optimized ELM Prediction Constraints | |
| CN111832832B (en) | District self-inspection system based on thing networking | |
| Yan et al. | A New Method for Anomaly Detection and Diagnosis of Ocean Observation System based on Deep Learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |