Method for improving energy efficiency and optimizing resource allocation of integrated energy system

Abstract

The invention belongs to the technical field of energy efficiency improvement and resource allocation, and discloses an energy efficiency improvement and resource allocation optimization method for an integrated energy system, and the method comprises the following specific use steps: S1, energy system evaluation and quantification target setting; s2, making a technology upgrading and multi-source complementation strategy; s3, investment analysis and system deployment implementation; S4, real-time monitoring and dynamic optimization scheduling; s5, an energy efficiency evaluation and continuous improvement mechanism: quantifying index kernel income, and regularly evaluating new technology iteration; s6, digital twinning and user side optimization; and S7, performing case verification and large-scale popularization. According to the invention, an energy efficiency evaluation and continuous improvement mechanism and digital twinning and user side optimization are matched for use; through digital twinning and user side optimization, the characteristics and problems of an energy system are comprehensively known through virtual simulation, and a basis is provided for optimizing a scheduling strategy; and the optimal energy efficiency scheme is automatically searched through reinforcement learning, and the energy utilization efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of energy efficiency improvement and resource allocation improvement, and specifically relates to a method for improving the energy efficiency and optimizing the resource allocation of an integrated energy system. Background Art

[0002] The method for improving the energy efficiency and optimizing the resource allocation of an integrated energy system is a scientific means and strategy adopted to achieve the efficient operation of the integrated energy system and the rational allocation of resources. In terms of energy efficiency improvement, it focuses on enhancing the utilization efficiency of the entire energy chain. Taking the integrated energy system of an industrial park containing multiple energy forms as an example, by means of advanced energy-saving equipment and technologies such as high-efficiency gas turbines and intelligent transformers, energy conversion and transmission losses are reduced. At the same time, using an intelligent control system, according to the real-time energy supply and demand situation, the energy production and distribution are accurately regulated to prevent energy waste, just as precisely as a household intelligent appliance control system adjusts the light brightness according to demand. In terms of optimizing resource allocation, this method rationally allocates and collaboratively optimizes multiple energies according to different energy characteristics, costs, and user demand preferences. For example, during the peak electricity consumption period in summer, the system preferentially uses renewable energy for power generation according to the real-time electricity price and energy supply, and converts the excess electric energy into heat energy for storage, and releases it during the low electricity consumption period or the peak heat energy demand period, just like a shrewd housekeeper reasonably arranges the procurement and usage plan to achieve the goal of reducing costs and improving benefits. In the current traditional methods in the field of energy management during use, they overly focus on improving the energy efficiency at the input end, but neglect the in-depth integration of renewable energy, which makes the green transformation of the energy structure extremely difficult, and the proportion of clean energy is far from the requirements of the sustainable development goal, which is not conducive to the long-term green development of the energy industry. At the same time, the dispatching control system reacts slowly, unable to timely capture the fluctuations in the energy market supply and demand, and it is also difficult to keep up with the changes in the dynamic demands of end-use energy equipment, thereby resulting in low energy allocation efficiency, poor utilization benefits, and relatively serious energy waste. Moreover, the system architecture design lacks flexibility, has weak adaptability to the special needs of different user scenarios, and cannot effectively balance the energy consumption differences among multiple scenarios such as industrial, commercial, and civilian, restricting the precise supply of energy services and squeezing the space for personalized optimization. These problems seriously restrict the improvement of energy management level and hinder the development of the energy industry towards the direction of high efficiency, greenness, and intelligence, and urgently need innovative methods to solve them. Summary of the Invention

[0003] The purpose of the present invention is to provide a method for improving the energy efficiency and optimizing the resource allocation of an integrated energy system to solve the problems raised in the above background art.

[0004] To achieve the above purpose, the present invention provides the following technical solution: A method for improving the energy efficiency and optimizing the resource allocation of an integrated energy system. The specific usage steps of this method are as follows:

[0005] S1: Energy System Assessment and Quantitative Target Setting: Conduct a digital inventory of energy assets and energy efficiency grading, scan QR codes or RFID to enter information to build a ledger, grade energy efficiency and generate heat maps, and retrofit inefficient equipment. Data is then collected, and algorithms are used to identify waste points and generate reports. Finally, policy-linked quantitative targets are set, carbon emission quotas are allocated to equipment, and dynamic updates are made according to the PDCA model to ensure compliance.

[0006] S2: Technology Upgrade and Multi-Source Complementary Strategy Development: Promote equipment iteration and upgrades, deploy high-efficiency energy-saving equipment and transform existing equipment, and clarify waste heat recovery efficiency targets; then build an energy coupling system, design a CCHP system, and match distributed photovoltaics with energy storage; finally, use LSTM neural networks to predict energy loads, dynamically adjust scheduling plans, set accuracy targets, and ensure precise matching of energy supply and demand;

[0007] S3: Investment Analysis and System Deployment: Conduct economic calculations and use sensitivity analysis models to assess the impact of energy price fluctuations on return on investment. Then, proceed with hardware deployment and build a monitoring and control network. Finally, develop and integrate software to achieve real-time monitoring of energy flows.

[0008] S4: Real-time Monitoring and Dynamic Optimization: Leveraging the IoT platform to deploy a LoRaWAN sensor network, comprehensive data collection on temperature, pressure, and flow rates across the campus is generated. This provides an information foundation for analysis, early warning, and scheduling, enabling real-time monitoring of energy system status. Energy efficiency early warning thresholds are set, with automatic alerts and recommendations issued when the load exceeds 80%. Algorithms combined with weather data dynamically adjust strategies to ensure a stable and efficient energy supply.

[0009] S5: Energy efficiency assessment and continuous improvement mechanism: Quantify the effectiveness by calculating the energy consumption reduction rate per unit of GDP, calculate the investment payback period to verify the economic feasibility, and regularly evaluate the feasibility of new technology integration to iteratively update the model;

[0010] S6: Digital Twin and User-Side Optimization: Conduct virtual simulations and build digital twin models to comprehensively simulate different scenarios of energy system load changes and policy changes affecting energy supply and demand during high summer temperatures. Accurately simulate system responses and gain a deeper understanding of characteristics and potential issues. Then, apply reinforcement learning optimization and machine learning algorithms to automatically optimize scheduling strategies based on simulation data to find the optimal energy efficiency solution. Finally, promote demand response and peak-valley electricity prices to increase participation among industrial users, balance energy supply and demand, and reduce peak pressure.

[0011] S7: Case verification and large-scale promotion: Deploy PV, storage and charging in industrial parks to reduce costs and increase efficiency, coordinate policies, and output a standard "source-grid-load-storage" solution.

[0012] Preferably, the specific steps for energy system assessment and quantitative target setting in S1 are as follows:

[0013] Step 1: Digital inventory of energy assets and energy efficiency grading

[0014] Enter the motor model and boiler capacity equipment information through scanning codes or RFID technology, build a digital ledger containing ID, location, and age; classify the energy efficiency of the equipment according to national standards, and mark the proportion of high-efficiency motors above IE3; generate an energy efficiency heat map, mark low-energy efficiency equipment in red, and give priority to including it in the energy-saving renovation plan;

[0015] Step 2: Data-driven energy consumption diagnosis and identification of waste points

[0016] Collect historical data from smart meters and flow meters, use clustering algorithms to identify peak periods and idle equipment, calculate the waste volume, and generate an energy-saving report;

[0017] The formula for the clustering algorithm is as follows:

[0018] Calculation of the clustering center, where where C k is the k-th clustering center, and S k is the sample set in the k-th clustering;

[0019] Step 3: Setting of policy-linked quantitative targets

[0020] Refer to the "Administrative Measures for Carbon Emission Trading" to allocate carbon emission quotas to specific equipment, and use the PDCA model to set dual targets: a 15% reduction in energy consumption per unit of output value and a 30% share of renewable energy; dynamically update the target values quarterly according to policy adjustments to ensure compliance.

[0021] Preferably, the specific usage steps of formulating the technology upgrade and multi-source complementary strategy in S2 are as follows:

[0022] Step 1: Equipment iterative upgrade

[0023] Deploy high-efficiency energy-saving equipment, including permanent magnet synchronous motors and absorption heat pumps, and at the same time transform existing equipment, install waste heat recovery devices, and clarify the target for improving waste heat recovery efficiency;

[0024] Step 2: Construction of an energy coupling system

[0025] Design a CCHP system, reasonably match distributed photovoltaics and energy storage systems, configure a lithium iron phosphate energy storage system, and use the energy storage system to smooth the fluctuations of photovoltaic power generation to achieve effective coupling and collaborative utilization of various energy forms;

[0026] Step 3: Achieving precise supply-demand matching

[0027] By using LSTM neural network technology to predict energy load, dynamically adjusting energy dispatch plans based on the prediction results, and setting targets for improving prediction accuracy, the park's prediction accuracy has increased to 92%, thereby ensuring an accurate match between energy supply and demand.

[0028] Preferably, the specific steps of investment analysis and system deployment implementation in S3 are as follows:

[0029] Step 1: Conduct economic calculations

[0030] Utilize sensitivity analysis models to focus on energy price fluctuations and assess their impact on investment returns; identify key energy types, including natural gas, and set price fluctuation ranges; accurately calculate returns under different volatility scenarios to provide economic feasibility for investment decisions;

[0031] Step 2: Promote hardware deployment

[0032] Install smart meters, temperature and humidity sensor monitoring equipment, and configure energy storage battery packs. Use advanced energy management systems to achieve inter-device linkage control, ensure the coordinated operation of various hardware devices, and build an efficient and intelligent hardware monitoring and control network.

[0033] Step 3: Software development and integration

[0034] Build an energy management system and integrate it with the SCADA platform; use the system to monitor energy flows in real time and set strict system response time standards.

[0035] Preferably, the specific steps of real-time monitoring and dynamic optimization scheduling in S4 are as follows:

[0036] Step 1: Comprehensive data collection and monitoring

[0037] Leveraging the IoT platform, we deploy sensor networks such as LoRaWAN to comprehensively collect multi-dimensional data on temperature, pressure, and flow within the park, ensuring park-wide data coverage. This data provides a rich and accurate foundation for subsequent analysis, early warning, and scheduling, enabling real-time monitoring of the energy system's operating status.

[0038] Step 2: Intelligent early warning and dynamic scheduling

[0039] An energy efficiency warning threshold is set to automatically trigger an alarm when the transformer load rate exceeds 80%, and targeted maintenance recommendations are pushed. At the same time, a model predictive control algorithm is used in combination with weather forecast data to dynamically adjust equipment operation strategies. When the photovoltaic output is insufficient, it automatically switches to the gas turbine to ensure the stability and efficiency of energy supply.

[0040] Preferably, the energy efficiency evaluation and continuous improvement mechanism in S5 refers to that in the process of energy management and sustainable development, index quantification is a key step. By calculating the decline rate of energy consumption per unit of GDP and the core index of carbon emission intensity, including a 18% reduction in carbon emissions achieved by the project, the energy conservation and emission reduction effects can be clearly presented; economic verification is also indispensable. Calculate the investment payback period. For example, the payback period of the integrated photovoltaic energy storage and charging system is about 5.2 years, and compare it with the income from carbon trading to provide a basis for project decision-making; in addition, new technologies need to be evaluated regularly, including the integration feasibility of hydrogen energy storage and virtual power plants, and the model needs to be updated and optimized in a timely manner.

[0041] Preferably, the specific usage steps of digital twin and user-side optimization in S6 are as follows:

[0042] Step 1: Conduct virtual simulation

[0043] Build a digital twin model to comprehensively simulate different scenarios, such as the load changes of the energy system and the impact of policy changes on energy supply and demand in summer high temperatures; through accurate simulation of the system's response in different scenarios, deeply understand the system characteristics and potential problems;

[0044] Step 2: Optimize using reinforcement learning

[0045] With the help of machine learning algorithms, based on the data obtained from virtual simulation, automatically optimize the scheduling strategy; let the algorithm continuously learn and adjust to find a scheduling plan that can achieve the optimal energy efficiency in different scenarios;

[0046] Q(s,a)←Q(s,a)+a[r+ymax a Q(s,a)-Q(s,a)], where s is the state, a is the action, r is the reward, α is the learning rate, and γ is the discount factor;

[0047] Step 3: Promote demand response

[0048] Vigorously promote the peak-valley electricity price mechanism, publicize the benefits and incentive policies of off-peak electricity consumption to users; for industrial users, which are the key group, take targeted measures to improve their participation rate, ensure that the participation rate of industrial users in demand response exceeds 60%, effectively balance energy supply and demand, and reduce the energy pressure during peak hours.

[0049] Preferably, the case verification and large-scale promotion in S7 refer to that the deployment of the integrated photovoltaic energy storage and charging system in the industrial park has achieved remarkable results, with the annual energy consumption cost reduced by 22% and the internal rate of return stabilized above 8%, achieving a win-win situation of economic and environmental benefits; on this basis, the park combines the "dual-carbon" goal, preferentially allocates the proportion of green electricity and actively participates in the carbon trading market to further promote green development; at the same time, the park pays attention to summarizing experience, refining a standardized solution, forming a replicable "source-network-load-storage" coordination template, and exporting it to similar parks or enterprises.

[0050] The beneficial effects of the present invention are as follows:

[0051] The present invention uses an energy efficiency evaluation and continuous improvement mechanism in combination with digital twin and user-side optimization; through digital twin and user-side optimization, a comprehensive understanding of the characteristics and problems of the energy system is achieved through virtual simulation to provide a basis for optimizing the dispatching strategy; reinforcement learning is used to automatically find the optimal energy efficiency solution to improve energy utilization efficiency; the promotion of demand response is advanced to encourage users to shift electricity consumption during off-peak hours, balance energy supply and demand, reduce the pressure during peak hours, and present the energy conservation and emission reduction effects in terms of indicators through the energy efficiency evaluation and continuous improvement mechanism, such as calculating the decline rate of energy consumption per unit of GDP and carbon emission intensity; economic verification provides a basis for project decision-making, calculates the investment payback period and compares it with the income from carbon trading; the feasibility of integrating new technologies is regularly evaluated, and the optimization model is updated in a timely manner; the combination of the two can accurately grasp the energy management effect, ensure the economic feasibility of the project, keep up with the technological forefront, promote the continuous upgrading and optimization of the energy system, and contribute to the sustainable development of energy. Brief Description of the Drawings

[0052] Figure 1 It is a schematic diagram of the overall usage process of the present invention. Specific Embodiments

[0053] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without making creative efforts shall fall within the protection scope of the present invention.

[0054] As Figure 1 shown, the embodiments of the present invention provide a method for improving the energy efficiency and optimizing the resource allocation of an integrated energy system. The specific usage steps of this method are as follows:

[0055] S1: Energy system evaluation and quantification target setting: Conduct digital inventory and energy efficiency grading of energy assets, scan codes or use RFID to enter information to build a ledger, perform energy efficiency grading and generate a heat map, and transform low-energy-consuming equipment; then collect data, use algorithms to identify waste points and generate reports; finally, set policy-linked quantification targets, allocate carbon emission quotas to equipment, and dynamically update according to the PDCA model to ensure compliance;

[0056] S2: Technology upgrade and multi-source complementary strategy formulation: Promote the iterative upgrade of equipment, deploy high-efficiency energy-saving equipment and transform existing equipment, and clarify the waste heat recovery efficiency target; then build an energy coupling system, design a CCHP system, and match distributed photovoltaics with energy storage; finally, use the LSTM neural network to predict the energy load, dynamically adjust the dispatching plan, set the accuracy target, and ensure the accurate matching of energy supply and demand;

[0057] S3: Investment Analysis and System Deployment: Conduct economic calculations and use sensitivity analysis models to assess the impact of energy price fluctuations on return on investment. Then, proceed with hardware deployment and build a monitoring and control network. Finally, develop and integrate software to achieve real-time monitoring of energy flows.

[0058] S4: Real-time Monitoring and Dynamic Optimization: Leveraging the IoT platform to deploy a LoRaWAN sensor network, comprehensive data collection on temperature, pressure, and flow rates across the campus is generated. This provides an information foundation for analysis, early warning, and scheduling, enabling real-time monitoring of energy system status. Energy efficiency early warning thresholds are set, with automatic alerts and recommendations issued when the load exceeds 80%. Algorithms combined with weather data dynamically adjust strategies to ensure a stable and efficient energy supply.

[0059] S5: Energy efficiency assessment and continuous improvement mechanism: Quantify the effectiveness by calculating the energy consumption reduction rate per unit of GDP, calculate the investment payback period to verify the economic feasibility, and regularly evaluate the feasibility of new technology integration to iteratively update the model;

[0060] S6: Digital Twin and User-Side Optimization: Conduct virtual simulations and build digital twin models to comprehensively simulate different scenarios of energy system load changes and policy changes affecting energy supply and demand during high summer temperatures. Accurately simulate system responses and gain a deeper understanding of characteristics and potential issues. Then, apply reinforcement learning optimization and machine learning algorithms to automatically optimize scheduling strategies based on simulation data to find the optimal energy efficiency solution. Finally, promote demand response and peak-valley electricity prices to increase participation among industrial users, balance energy supply and demand, and reduce peak pressure.

[0061] S7: Case verification and large-scale promotion: Deploy PV, storage and charging in industrial parks to reduce costs and increase efficiency, coordinate policies, and output a standard "source-grid-load-storage" solution.

[0062] The specific steps for energy system assessment and quantitative target setting in S1 are as follows:

[0063] Step 1: Digital inventory of energy assets and energy efficiency grading

[0064] By scanning codes or using RFID technology to enter equipment information such as motor model and boiler capacity, a digital ledger containing ID, location, and age is constructed. Equipment is graded for energy efficiency according to national standards such as GB18613-2020, noting the proportion of high-efficiency motors rated IE3 and above. An energy efficiency heat map is generated, with low-efficiency equipment, such as air compressors with energy efficiency ratings below 3, marked in red, and prioritized for inclusion in energy-saving renovation plans.

[0065] Step 2: Data-driven energy consumption diagnosis and waste point identification

[0066] Collect historical data of smart meters and flow meters with a 15-minute granularity and a 12-month cycle. Use clustering algorithms to identify peak periods and idle equipment, calculate the waste amount (idle power × time × electricity price), and generate an energy-saving report.

[0067] The formula for the clustering algorithm is as follows:

[0068] Calculation of the clustering center, where where C k is the k-th clustering center, and S k is the set of samples in the k-th cluster;

[0069] Step 3: Setting of policy-linked quantitative targets

[0070] Refer to the "Administrative Measures for Carbon Emission Trading" to allocate carbon emission quotas to specific equipment. For example, gas boilers account for 25%. Use the PDCA model to set dual targets: the energy consumption per unit output value is reduced by 15%, such as from 0.3 tons of standard coal per 10,000 yuan to 0.255 tons, and the proportion of renewable energy reaches 30% (photovoltaic + biomass energy 28% + 2%). Update the target values dynamically every quarter according to policy adjustments such as tightened carbon quotas to ensure compliance.

[0071] This solution realizes precise management and efficiency improvement through digital ledgers and energy efficiency grading (such as marking the proportion of IE3 motors), can quickly locate low-energy equipment (such as marking air compressors in red), increasing the pertinence of energy-saving transformation by more than 40% and optimizing resource allocation; based on the clustering analysis of 15-minute energy consumption data, accurately identify peak electricity consumption periods and idle equipment (such as an air compressor with an idle rate of 42%), quantify energy waste and generate an energy-saving report, with an annual electricity cost savings of up to 87,000 yuan, helping to reduce costs driven by data; allocate carbon emission quotas to specific equipment (such as gas boilers accounting for 25%), combine with PDCA target decomposition to ensure compliance with policies such as a 15% reduction in energy consumption per unit output value and a 30% proportion of renewable energy, and the dynamic calibration mechanism can flexibly respond to changes such as tightened carbon quotas, realizing policy-linked risk avoidance; the dual target system (energy efficiency + policy) promotes the green transformation of enterprises. For example, an enterprise reduces carbon emissions by 18% through this solution and obtains a government subsidy of 320,000 yuan, achieving a win-win situation of economic and environmental benefits and empowering sustainable development.

[0072] Among them, the specific usage steps of the technology upgrade and multi-source complementary strategy formulation in S2 are as follows:

[0073] Step 1: Equipment iteration and upgrade

[0074] Deploy high-efficiency energy-saving equipment, including permanent magnet synchronous motors and absorption heat pumps. At the same time, retrofit existing equipment and install waste heat recovery devices, and clarify the target for improving waste heat recovery efficiency, such as increasing the flue gas waste heat recovery efficiency by 12% - 15%, so as to improve energy utilization efficiency;

[0075] Step 2: Construct an energy coupling system

[0076] Design a CCHP (Combined Cooling, Heating and Power) system, reasonably match the distributed photovoltaic and energy storage systems, configure a lithium iron phosphate energy storage system, use the energy storage system to smooth the fluctuations of photovoltaic power generation, and achieve the effective coupling and collaborative utilization of various energy forms;

[0077] Step 3: Achieve precise supply-demand matching

[0078] Use the LSTM (Long Short-Term Memory) neural network technology to predict the energy load, dynamically adjust the energy scheduling plan according to the prediction results, and set the goal of improving the prediction accuracy, such as increasing the prediction accuracy of a certain park to 92%, so as to ensure the precise matching of energy supply and demand.

[0079] At the equipment level, through iterative upgrade and deployment of high-efficiency energy-saving equipment such as permanent magnet synchronous motors and absorption heat pumps, and installation of waste heat recovery devices, and clarifying goals such as increasing the flue gas waste heat recovery efficiency by 12%-15%, it can greatly improve the energy utilization efficiency, reduce the equipment energy consumption, and reduce the enterprise operation cost.

[0080] In terms of energy coupling, build a CCHP system, reasonably match the distributed photovoltaic and lithium iron phosphate energy storage systems, use the energy storage to smooth the photovoltaic fluctuations, and achieve the effective coupling and collaborative utilization of various energies. This not only enhances the stability and reliability of energy supply, but also can optimize the energy structure, increase the consumption ratio of renewable energy, and promote the transformation of energy to green and low-carbon.

[0081] In the supply-demand matching link, use the LSTM neural network to predict the load, dynamically adjust the energy scheduling plan, set the goal of improving the prediction accuracy, such as increasing the prediction accuracy of the park to 92%, which can ensure the precise matching of energy supply and demand, avoid energy waste, and further improve the comprehensive benefit of energy utilization.

[0082] Among them, the specific steps of investment analysis and system deployment implementation in S3 are as follows:

[0083] Step 1: Conduct economic calculation

[0084] Use the sensitivity analysis model, focus on the energy price fluctuation factor, evaluate its impact on the internal rate of return (IRR) of investment; clarify the key energy varieties, such as natural gas, set the price fluctuation range, for example, ±10%, and accurately calculate the income changes under different fluctuation scenarios, providing an economic feasibility basis for investment decisions;

[0085] Step 2: Promote the hardware deployment work

[0086] Install monitoring devices such as smart electricity meters and temperature and humidity sensors, and configure energy storage battery packs at the same time; adopt an advanced Energy Management System (EMS) to achieve linkage control among devices, ensure the coordinated operation of each hardware device, and build an efficient and intelligent hardware monitoring and regulation network;

[0087] Step 3: Carry out software development and integration

[0088] Build an Energy Management System (EMS) and integrate it with the SCADA platform; realize real-time monitoring of the energy flow through this system, and set strict system response time standards; for example, the response time of a certain system needs to be less than 1 second to ensure the timeliness and accuracy of energy management.

[0089] In terms of economic calculation, use a sensitivity analysis model to focus on the impact of energy price fluctuations on the return on investment, identify key energy varieties and set the fluctuation range, and accurately calculate the income changes under different scenarios. This enables investors to clearly understand the economic feasibility of the project under different energy price conditions, make preparations for risk response in advance, avoid investment losses caused by energy price fluctuations, and improve the scientificity and accuracy of investment decisions.

[0090] In the hardware deployment work, install various monitoring devices and energy storage battery packs, and achieve device linkage control through EMS. This builds an efficient and intelligent hardware monitoring and regulation network, which can master the energy usage situation in real time, optimize energy distribution, improve energy utilization efficiency, and reduce energy waste.

[0091] In the software development and integration link, build EMS and integrate it with the SCADA platform to achieve real-time monitoring of the energy flow and set strict response time standards. This ensures the timeliness and accuracy of energy management, can quickly discover and solve problems in energy use, ensure the stable operation of the energy system, and improve the overall energy management level.

[0092] Among them, the specific steps of real-time monitoring and dynamic optimization scheduling in S4 are as follows:

[0093] Step 1: Comprehensive data collection and monitoring

[0094] With the help of the IoT platform, deploy a LoRaWAN sensor network to achieve comprehensive collection of multi-dimensional data such as temperature, pressure, and flow in the park, ensuring that the data covers the entire park; through these data, provide a rich and accurate information basis for subsequent analysis, early warning, and scheduling, so as to master the operation status of the energy system in real time;

[0095] Step 2: Intelligent early warning and dynamic scheduling

[0096] Set the energy efficiency warning threshold. For example, when the transformer load rate exceeds 80%, an alarm is automatically triggered, and targeted maintenance suggestions are pushed. At the same time, adopt the model predictive control (MPC) algorithm, combine with weather prediction data, and dynamically adjust the equipment operation strategy. When the photovoltaic power generation is insufficient, automatically switch to the gas turbine to ensure the stability and efficiency of energy supply.

[0097] Comprehensive data collection and monitoring, as well as intelligent early warning and dynamic scheduling, bring many benefits. Comprehensive collection of multi-dimensional data can accurately grasp the operation status of the energy system in real time, provide rich and accurate basis for subsequent decision-making, and help to discover potential problems in a timely manner. Setting the energy efficiency warning threshold can alarm in time and push maintenance suggestions when the equipment load is abnormal, avoiding equipment damage and energy waste. Adopting the MPC algorithm combined with weather prediction to dynamically adjust the equipment operation strategy can flexibly switch the energy supply mode according to different situations, ensure the stable and efficient energy supply, reduce the operation cost, and improve the energy utilization efficiency.

[0098] Among them, the energy efficiency evaluation and continuous improvement mechanism in S5 refers to that in the process of energy management and sustainable development, index quantification is a key step. By calculating the core indicators such as the decline rate of energy consumption per unit GDP and carbon emission intensity, including a project achieving a 18% reduction in carbon emissions, the energy conservation and emission reduction effects can be clearly presented. Economic verification is also indispensable. Calculate the investment payback period, such as the payback period of the integrated photovoltaic energy storage and charging system is about 5.2 years, and compare it with the carbon trading income to provide a basis for project decision-making. In addition, new technologies need to be evaluated regularly, including the integration feasibility of hydrogen energy storage and virtual power plants, and the model needs to be updated and optimized in a timely manner. In this way, the energy management effect can be accurately grasped, the economic feasibility of the project can be ensured, keep up with the technological forefront, and promote the continuous upgrade and optimization of the energy system.

[0099] Among them, the specific use steps of digital twin and user-side optimization in S6 are as follows:

[0100] Step 1: Carry out virtual simulation

[0101] Build a digital twin model to comprehensively simulate different scenarios, such as the load changes of the energy system and the impact of policy changes on energy supply and demand in high temperature in summer. Through accurate simulation of the system's response in different scenarios, deeply understand the system characteristics and potential problems, and provide data support and theoretical basis for subsequent optimization of the dispatching strategy.

[0102] Step 2: Optimize using reinforcement learning

[0103] With the help of machine learning algorithms, based on the data obtained from virtual simulation, automatically optimize the dispatching strategy. Let the algorithm continuously learn and adjust to find the dispatching plan that can achieve the optimal energy efficiency in different scenarios.

[0104] The reinforcement learning optimization such as the Q-learning formula is as follows;

[0105] Q(s,a) ← Q(s,a) + α[r + γ max a Q(s,a) - Q(s,a)], where s is the state, a is the action, r is the reward, α is the learning rate, and γ is the discount factor;

[0106] Step 3: Promote demand response

[0107] Vigorously promote the peak-valley electricity price mechanism and publicize the benefits and incentive policies of off-peak electricity consumption to users; for key groups such as industrial users, take targeted measures to improve their participation rate, ensure that the participation rate of industrial users in demand response exceeds 60%, effectively balance energy supply and demand, and reduce the energy pressure during peak hours.

[0108] Carrying out digital twins and user-side optimization brings many benefits. Through virtual simulation, the characteristics and problems of the energy system in different scenarios can be comprehensively understood, providing a reliable basis for optimizing the dispatching strategy. Using reinforcement learning to optimize the dispatching strategy, the algorithm can automatically find the optimal energy efficiency solution to improve energy utilization efficiency. Promoting demand response, using the peak-valley electricity price mechanism to encourage users to consume electricity off-peak, especially increasing the participation rate of industrial users, can effectively balance energy supply and demand, reduce the energy pressure during peak hours, reduce energy waste, improve the stability and economy of the energy system, and contribute to the sustainable development of energy.

[0109] Among them, the case verification and large-scale promotion in S7 refer to the remarkable results of deploying the optical storage and charging system in the industrial park, with the annual energy consumption cost reduced by 22% and the internal rate of return stabilized above 8%, achieving a win-win situation of economic and environmental benefits; on this basis, the park combines the "dual carbon" goal, gives priority to configuring the proportion of green electricity and actively participates in the carbon trading market to further promote green development; at the same time, the park pays attention to summarizing experience, refining a standardized solution, forming a replicable "source-network-load-storage" coordination template, and exporting it to similar parks or enterprises. This not only helps to improve the energy utilization efficiency of the entire industry and reduce operating costs, but also provides green development ideas for more entities, accelerating the process of energy transformation and sustainable development.

[0110] The deployment of the optical storage and charging system in an industrial park has achieved remarkable results, with the annual energy consumption cost significantly reduced by 22% and the internal rate of return stabilized above 8%, achieving a double harvest of economic and environmental benefits. The park then combines the "dual carbon" goal, increases the proportion of green electricity configuration, actively participates in carbon trading, and deepens green development. Moreover, the park is good at summarizing experience, refining and exporting a standardized "source-network-load-storage" coordination template. This measure is of great significance, which can not only improve the energy utilization efficiency of the industry and reduce operating costs, but also provide a green development example for other entities, strongly promoting energy transformation and accelerating the pace of sustainable development.

[0111] It should be noted that in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variation thereof are intended to cover non-exclusive inclusion, such that a process, method, article or apparatus comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or elements inherent to such process, method, article or apparatus.

[0112] 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 principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. Method for improving energy efficiency and optimizing resource allocation in integrated energy systems, characterized in that: The specific usage steps of this method are as follows: S1: Energy system assessment and quantification target setting: Conduct digital inventory of energy assets and energy efficiency grading, scan codes or use RFID to input information to build a ledger, conduct energy efficiency grading and generate a heat map, and transform low-efficiency equipment; Then collect data, use algorithms to identify waste points and generate reports; Finally, set policy-linked quantification targets, allocate carbon emission quotas to equipment, and dynamically update according to the PDCA model to ensure compliance; S2: Technology upgrade and multi-source complementary strategy formulation: Promote equipment iteration and upgrade, deploy energy-efficient equipment and transform existing equipment to improve waste heat recovery efficiency; Build an energy coupling system, use the LSTM neural network to predict load, and dynamically optimize the scheduling plan; S3: Investment analysis and system deployment implementation: Conduct economic calculations, use the sensitivity analysis model to evaluate the impact of energy price fluctuations on the return on investment; Then promote hardware deployment and build a monitoring and control network; Finally, carry out software development and integration to achieve real-time monitoring of the energy flow; S4: Real-time monitoring and dynamic optimization scheduling: Deploy a LoRaWAN sensor network through the IoT platform, comprehensively collect temperature, pressure, and flow data in the park, grasp the status of the energy system in real time, set the energy efficiency warning threshold, automatically alarm and push suggestions when the load rate exceeds 80%, and dynamically adjust the strategy in combination with the weather to ensure stable and efficient energy supply; S5: Energy efficiency assessment and continuous improvement mechanism: Quantify the effectiveness by calculating the decline rate of energy consumption per unit of GDP, calculate the payback period to verify the economy, and regularly evaluate the feasibility of new technology integration to iterate and update the model; S6: Digital twin and user-side optimization: Build a digital twin model through virtual simulation, simulate the energy system scenarios under high summer temperatures and policy changes, and accurately grasp the response characteristics and problems; Use reinforcement learning to optimize scheduling to obtain the optimal energy efficiency plan; Promote demand response, publicize peak-valley electricity prices, improve the participation of industrial users, and balance supply and demand; S7: Case verification and large-scale promotion: The park closely adheres to the "dual carbon" goal, increases the proportion of green power configuration, and actively participates in carbon trading; In addition, the park has refined a standardized "source-network-load-storage" coordination template to provide replicable and referenceable solutions for similar parks or enterprises.

2. The method for improving the energy efficiency and optimizing the resource allocation of the integrated energy system according to claim 1, wherein: The specific usage steps of energy system assessment and quantification target setting in S1 are as follows: Step 1: Digital inventory of energy assets and energy efficiency grading Input equipment information such as motor model and boiler capacity through scanning codes or RFID technology to build a digital ledger containing ID, location, and age; Conduct energy efficiency grading of equipment according to national standards, and mark the proportion of high-efficiency motors above IE3; Generate an energy efficiency heat map, mark low-efficiency equipment in red, and give priority to including it in the energy-saving transformation plan; Step 2: Data-driven energy consumption diagnosis and waste point identification Collect historical data of smart meters and flow meters, use clustering algorithms to identify peak periods and unloaded equipment, calculate the waste volume, and generate an energy-saving report; The formula of the clustering algorithm is as follows: Calculation of cluster center, where where C k is the k-th cluster center, and S k is the set of samples in the k-th cluster; Step 3: Policy-linked quantification target setting Allocate carbon emission quotas to specific equipment, and use the PDCA model to set dual targets: Dynamically update the target values according to policy adjustments every quarter to ensure compliance.

3. The method for improving the energy efficiency and optimizing the resource allocation of the integrated energy system according to claim 1, wherein: The specific usage steps for technology upgrade and multi-source complementary strategy formulation in S2 are as follows: Step 1: Equipment iterative upgrade Deploy energy-efficient equipment, including permanent magnet synchronous motors and absorption heat pumps. At the same time, retrofit existing equipment by installing waste heat recovery devices and clarify the target for improving waste heat recovery efficiency; Step 2: Construct an energy coupling system Design a CCHP system, reasonably match distributed photovoltaic and energy storage systems, configure a lithium iron phosphate energy storage system, and use the energy storage system to smooth the fluctuations of photovoltaic power generation to achieve effective coupling and collaborative utilization of various energy forms; Step 3: Achieve precise supply-demand matching Use the LSTM neural network technology to predict the energy load, dynamically adjust the energy dispatching plan according to the prediction results, set the target for improving the prediction accuracy, and increase the prediction accuracy of the park to 92% to ensure the precise matching of energy supply and demand.

4. The method for improving the energy efficiency and optimizing the resource allocation of the integrated energy system according to claim 1, wherein: The specific steps for investment analysis and system deployment implementation in S3 are as follows: Step 1: Conduct economic calculations Use a sensitivity analysis model, focus on the factor of energy price fluctuations, evaluate its impact on the return on investment; clarify key energy varieties, including natural gas, and set the price fluctuation range; accurately calculate the income changes under different fluctuation scenarios to provide an economic feasibility basis for investment decisions; Step 2: Promote hardware deployment work Install intelligent electricity meters, temperature and humidity sensor monitoring devices, and at the same time configure energy storage battery packs; adopt an advanced energy management system to achieve linkage control between devices, ensure the coordinated operation of each hardware device, and build an efficient and intelligent hardware monitoring and control network; Step 3: Conduct software development and integration Build an energy management system and integrate it with the SCADA platform; realize real-time monitoring of the energy flow through this system and set strict system response time standards.

5. The method for improving the energy efficiency and optimizing the resource allocation of the integrated energy system according to claim 1, wherein: The specific steps for real-time monitoring and dynamic optimization dispatching in S4 are as follows: Step 1: Comprehensive data collection and monitoring With the help of the IoT platform, deploy a sensor network such as LoRaWAN to achieve comprehensive collection of multi-dimensional data such as temperature, pressure, and flow in the park, ensuring that the data covers the entire park; through these data, provide a rich and accurate information basis for subsequent analysis, early warning, and dispatching, and facilitate real-time grasping of the operating status of the energy system; Step 2: Intelligent early warning and dynamic dispatching Set the energy efficiency early warning threshold, automatically trigger an alarm when the transformer load rate exceeds 80%, and push targeted maintenance suggestions; at the same time, adopt a model predictive control algorithm, combined with weather prediction data, to dynamically adjust the device operation strategy; when the photovoltaic output is insufficient, automatically switch to a gas turbine to ensure the stability and efficiency of energy supply.

6. The method for improving the energy efficiency and optimizing the resource allocation of the integrated energy system according to claim 1, wherein: The energy efficiency assessment and continuous improvement mechanism in S5 refers to the fact that indicator quantification is a key step in the process of energy management and sustainable development. By calculating the energy consumption reduction rate per unit GDP and the core indicators of carbon emission intensity, including the 18% reduction in carbon emissions achieved by the project, the energy conservation and emission reduction results can be clearly presented. Economic verification is also indispensable. The investment payback period is calculated. For example, the payback period of the integrated photovoltaic storage and charging system is about 5.2 years, and compared with the carbon trading income to provide a basis for project decision-making. In addition, new technologies need to be regularly evaluated, including the integration feasibility of hydrogen energy storage and virtual power plants, and the optimization model needs to be updated in a timely manner.

7. The method for improving the energy efficiency and optimizing the resource allocation of the integrated energy system according to claim 1, characterized in that: The specific steps for using digital twins and user-side optimization in S6 are as follows: Step 1: Conduct virtual simulation Build a digital twin model to comprehensively simulate different scenarios, including load changes in the energy system during high summer temperatures and the impact of policy changes on energy supply and demand. By accurately simulating the system's response in different scenarios, we can gain a deeper understanding of system characteristics and potential problems. Step 2: Optimize using reinforcement learning Leveraging machine learning algorithms, we automatically optimize scheduling strategies based on data obtained from virtual simulations. This allows the algorithm to continuously learn and adjust to find scheduling solutions that achieve optimal energy efficiency in different scenarios. The optimization formula using reinforcement learning is as follows; Q(s,a) ← Q(s,a) + α[r + γ max a Q(s',a') - Q(s,a)], where s is the state, a is the action, r is the reward, α is the learning rate, and γ is the discount factor; Step 3: Promote demand response Vigorously promote the peak-valley electricity price mechanism, and publicize the benefits and incentive policies of off-peak electricity consumption to users; take targeted measures to increase the participation of key groups of industrial users, ensure that the coverage rate of industrial users' participation in demand response exceeds 60%, effectively balance energy supply and demand, and reduce energy pressure during peak hours.

8. The method for improving the energy efficiency and optimizing the resource allocation of the integrated energy system according to claim 1, wherein: The case verification and large-scale promotion in S7 refers to the significant results achieved in deploying solar-storage-charging systems in industrial parks, which has achieved a win-win situation in terms of economic and environmental benefits. On this basis, the parks, in line with the "dual carbon" goals, prioritize the allocation of green electricity proportions and actively participate in the carbon trading market to further promote green development. At the same time, the parks focus on experience summary, extracting standardized solutions, and forming replicable "source-grid-load-storage" collaborative templates, which are exported to similar parks or enterprises.

Images