Big customer integrated energy online and offline full-period service method and system

Abstract

The invention belongs to the technical field of comprehensive energy service and energy storage, and relates to a big customer comprehensive energy online and offline full-cycle service method and system, the system and the method are driven by data, user load characteristics are automatically analyzed to identify peak-valley difference users, energy storage economic optimal capacity is calculated in physical and investment constraints by adopting a traversal algorithm, and energy storage efficiency is improved. An executable charging and discharging strategy is generated, automatic report generation and multi-terminal interaction are achieved, online and offline platforms are integrated, full-period service management from customer intention and technical analysis to operation is achieved, and the problems that enterprise user energy storage investment decision depends on experience, technical and economic analysis is complex, and the service process is split are solved. The technical effects of improving investment decision scientificity, ensuring global optimization of configuration, reducing energy consumption cost and optimizing service efficiency and customer experience are achieved.

Description

Technical Field

[0001] This application relates to the field of integrated energy services and energy storage technology, specifically a method and system for providing integrated energy services to large customers both online and offline throughout the entire lifecycle. Background Technology

[0002] As the energy transition deepens, more and more enterprise users hope to adopt integrated energy solutions to reduce energy costs and improve energy efficiency. However, enterprises face many challenges in the initial stages of integrated energy planning: First, there is the blindness in investment decisions: enterprise users lack professional analytical tools and find it difficult to accurately determine whether energy storage and other facilities are needed based on their own electricity load characteristics. Investment decisions often rely on experience, leading to high risks. Second, the technical and economic analysis is complex: the capacity configuration of energy storage systems is closely coupled with charging and discharging strategies, directly affecting the initial investment and long-term returns of the project. Manually calculating the optimal configuration and return model is very difficult and cannot guarantee scientific accuracy. Third, the service chain is broken: in traditional service models, customer management, technical solutions, investment assessment, and service implementation are disconnected, failing to provide enterprise users with a full-cycle, integrated service experience from intention to implementation and continuous operation. Finally, there is a lack of standardized tools: the market lacks standardized software platforms that can integrate customer management, technical analysis, economic assessment, and knowledge bases, resulting in low service efficiency and poor solution replicability. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this application provides a comprehensive online and offline full-cycle service method and system for large customers' integrated energy needs, in order to solve the problems mentioned above, such as enterprise users' reliance on experience in energy storage investment decisions, complex techno-economic analysis, and fragmented service processes.

[0004] To achieve the above objectives, this application provides the following technical solution: A comprehensive online and offline full-cycle energy service method for major customers includes the following steps: S1 user load analysis and type determination: Obtain electricity consumption data from enterprise users and generate time-series load data. A typical daily load power curve, in which t Calculate the peak-to-valley difference rate of the curve at the given time of day. Using formula Calculation, where P peak Maximum daily load P valley The minimum daily load, and compared with the preset threshold T threshold If a comparison is made, If so, the user is determined to be a peak-valley difference user suitable for configuring energy storage; S2 calculates the current year's basic electricity cost.C base Using formula Calculation, where P load ( t )for t Time-based load data, π( t )for t Time-of-use electricity price data for the user's location at any given time. Δt For time intervals, T This refers to the total number of time intervals after dividing the data according to a set time interval, and includes the unit capacity cost C. cap Charge / discharge efficiency η, cycle life N cycle S3 determines the energy storage system parameters; S3 determines the maximum configurable capacity of the energy storage system. E max , ,in E space Install space constraints for users. E transformer Due to user transformer capacity constraints, E budget To constrain user investment budgets; S4 uses preset step sizes ΔE As an increment, from the capacity value of 0 to... E max Iterate through the values, for i For each candidate capacity of the index number E i Calculate its initial investment cost I initial (E i ) It simulates its operation over its entire lifespan of N years and calculates the benefits including electricity cost savings. B saving (y) Annual net income from operating and maintenance costs B net (y, E i ) Then calculate the net present value of the capacity plan. NPV(E i ) ,in y For the year; S5 compares all candidate capacities E i Corresponding net present value NPV(E i ) Select the option that makes the net present value (NPV) NPV(E i ) Maximum capacityE opt As the optimal economic allocation solution, and based on E opt , P load (t) and π(t) Generate a time-granular charge / discharge strategy plan table containing a sequence of charge / discharge power commands. Schedule(t) The charging and discharging power command sequence is used to guide the energy storage system to perform automated peak shaving and valley filling operations; S6 based on E opt The system automatically generates an investment analysis report, including the payback period (PBP), internal rate of return (IRR), and visual charts, based on the corresponding cash flow forecast data. Preferably, the economically optimal capacity configuration traversal calculation in step S4 specifically includes: the initial investment cost... I initial ( E i The formula for calculating ) is: C fixed The net present value is a fixed cost independent of capacity. NPV(E i ) The calculation formula is: ,in r The discount rate is... N For the full lifecycle of financial analysis; annual net revenue from operating costs. The calculation formula is ,in for y The annual electricity cost savings are calculated using the following formula: ,in for y year t Discharge power at any given time Δt For time intervals, π peak ( t )for t Peak electricity price at any given time π valley ( t )for t Off-peak electricity prices at specific times; for y Annual battery replacement cost.

[0005] Preferably, the annual net revenue from operation and maintenance costs Use formula ,in for y The annual electricity subsidy income is calculated using the following formula: ,in for y Total annual discharge Ssub The subsidy amount is provided to the unit in the user's location. Based on the same inventive concept, this application also achieves this through the following technical solutions: A comprehensive online and offline full-lifecycle energy service system for large customers includes a data access and integration layer, a core analysis engine, a full-lifecycle service management platform, and a multi-terminal interaction and control system. The data access and integration layer is used to collect load time-series data through standardized communication protocols. P load (t) And store the time-of-use electricity pricing model. π(t) The system includes energy storage system parameters; the core analysis engine is connected to the data access and integration layer, and is used to perform calculations for load feature extraction, economic optimal configuration solution, and charging and discharging strategy generation; the full-cycle service management platform is integrated with the core analysis engine through an API interface, and is used to manage customer technical files, advance business opportunity processes, and automatically generate structured investment analysis reports. The business opportunity advancement is used to trigger the implementation of the large customer integrated energy online and offline full-cycle service method and the creation of offline service tasks as described in any one of claims 1 to 3; the multi-terminal interaction and control system is used to provide access and interaction interfaces for different levels of functions for offline service provider internal personnel, on-site engineering personnel, and enterprise users.

[0006] Preferably, the core analysis engine includes a load feature extractor, an economic optimal configuration solver, and a charging / discharging strategy generator; the load feature extractor is used to process load time-series data. P load (t) Process the data and calculate the peak-to-valley difference rate. R peak-valley The economic optimal allocation solver incorporates a traversal algorithm to address capacity limitations. E max Within the constraints, with step size ΔE Traverse and calculate all candidate capacities E i Net present value NPV(E i ) and output the globally optimal capacity. E opt The charge / discharge strategy generator is used to base on... E opt , P load (t) and π(t) Generate charge / discharge strategy plan table Schedule(t) .

[0007] Preferably, the full-cycle service management platform includes an intelligent business opportunity advancement pipeline module and a structured report generation and rendering service module; the intelligent business opportunity advancement pipeline is configured as a business opportunity state machine-driven process engine, used to automatically associate the changes in business opportunity state with the analysis tasks of the core analysis engine, so that user load analysis or economically optimal capacity configuration traversal calculation is automatically triggered during the business opportunity advancement process; the structured report generation and rendering service is used to call preset document templates and generate the report output by the core analysis engine. E opt , NPV and Schedule(t) The data is automatically populated and formatted for standard document output.

[0008] Preferably, the multi-terminal interaction and control system includes a mobile field service terminal, which supports offline and online modes, and is used to quickly import field load data and call the lightweight computing service of the core analysis engine for scheme simulation.

[0009] Preferably, it also includes a system integration and extension interface, which provides a standardized RESTful API interface for third-party systems to access data or trigger services. The third-party systems include enterprise ERP systems, power grid dispatching systems, or energy storage battery management systems (BMS).

[0010] Preferably, the full-cycle service management platform also includes a knowledge graph and case reasoning library, which stores anonymized historical project case data models. When analyzing new customers, it automatically retrieves and recommends similar historical cases and their configuration schemes based on the similarity of load characteristics.

[0011] Preferably, the multi-terminal interaction and control system includes a client-side mini-program or portal for enterprise users. This interface is not only used to display analysis reports and revenue forecasts to customers, but also serves as an entry point for customer-side data confirmation and a feedback channel for real-time operation data of the energy storage system.

[0012] Compared to existing technologies, the advantages of this solution are as follows: This application's comprehensive online and offline full-cycle energy service method and system for large customers is based on data-driven and algorithmic models. It provides a systematic solution that combines online analysis and calculation with offline professional services, offering enterprise users comprehensive energy services from planning to operation in a scientific and economical manner. It automates and models complex technical and economic analyses and deeply integrates them with business processes, achieving scientific, economical, and efficient comprehensive energy services. By acquiring enterprise electricity consumption data through the data access and integration layer and utilizing the feature calculation algorithm in the core analysis engine, it automatically identifies peak-valley characteristics in user load curves, accurately selecting "peak-valley difference users" suitable for energy storage optimization from massive amounts of data, replacing traditional inefficient manual experience-based judgment, thus constructing an automated user load characteristic analysis technology; and it designs and implements an economically optimal capacity configuration traversal algorithm. Within the technical boundaries defined by physical constraints such as installation space and transformer capacity, this algorithm traverses all possible configurations with a high-precision step size of 1 kWh. It integrates battery degradation models, operation and maintenance cost models, and electricity price policy models to simulate the entire life cycle cash flow, thereby outputting the global economic optimal solution under given constraints. This solves the technical problem of blind configuration in traditional schemes, thus realizing the quantitative calculation technology of economic optimal configuration of energy storage systems. Based on the optimal capacity configuration, combined with real-time or predicted load curves and time-of-use electricity price structures, a strategy optimization engine generates a time-granular sequence of charging and discharging control instructions. This strategy can be directly or after conversion and sent to the energy storage battery management system (BMS) to achieve automated operation control for peak shaving and valley filling, realizing the generation of executable dynamic charging and discharging strategy control technology. It integrates a data access and integration layer, a core analysis engine, a full-cycle service management platform, and a multi-terminal interaction and control system to build a physical system that supports online, data-driven, and standardized closed-loop management of the entire process from customer intention, customer management, technical analysis, project business negotiation to contract signing, operation, and knowledge accumulation. This platform exchanges data with power grid data systems, enterprise information systems, and equipment monitoring platforms through standardized API interfaces, forming a closed-loop online and offline technical service system. This creates an integrated, modular, full-lifecycle service technology platform. The solution also generates reusable knowledge assets, enabling customers to deeply participate in and transparently understand the entire project process, transforming them from one-time transactions into long-term partnerships. This improves customer satisfaction and loyalty, raising the customer satisfaction survey score from 7.5 to 9.2 out of 10. This proposed solution is a comprehensive solution that generates direct economic benefits, improves industry efficiency, reduces investment risks, and creates social value. It successfully transforms integrated energy services from a "craft" relying on personal experience into a standardized, scalable, and replicable modern service industry based on data and algorithms, providing solid technical support and a business model example for the high-quality development of the integrated energy industry. Attached Figure Description

[0013] Figure 1 This is a general flowchart of an embodiment of the method of this application; Figure 2 This is an architectural block diagram of one embodiment of the system in this application; Figure 3 The revenue-capacity relationship curve in the visualization chart generated by step S6 of an embodiment of the method of this application describes the economic optimization curve of energy storage capacity configuration, where the X-axis is the battery capacity, the Y-axis is the net revenue, and the peak point is the economically optimal configuration point. Detailed Implementation

[0014] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0015] This application provides a method for providing comprehensive online and offline full-cycle energy services to large customers. This embodiment includes the following steps: Step S1 mainly involves user load analysis and type determination. Specifically, it involves acquiring electricity consumption data from enterprise users and generating typical daily load power curves. ,in t In this embodiment, the time interval is 15 minutes, and the characteristic parameters of the curve are calculated, such as the peak-to-valley difference rate. , ,in P peak Maximum daily load P valley Select the peak-valley difference rate for the minimum daily load. R peak-valley Select strategy and match with preset threshold T threshold For comparison, this embodiment can set T threshold It is 40%, if R peak-valley >T threshold If the peak-valley difference is correctly identified, the user is deemed suitable for configuring energy storage and can proceed to the subsequent optimization process. The selection strategy can be based on comparing the highest peak-valley difference rate over a certain period, the average peak-valley difference rate, the peak-valley difference rate on a specific date, or other methods. Different selection strategies have different focuses; for example, the highest peak-valley difference rate focuses on extreme fluctuations, the average peak-valley difference rate focuses on typical daily fluctuations, and the peak-valley difference rate on a specific date focuses on data from a specific business scenario. This embodiment places greater emphasis on extreme fluctuations to ensure the capacity of the energy storage system. Characteristic parameters may also include load factor. ,in P averageThis represents the daily average load. Taking a manufacturing company as an example, the method in this embodiment calculates its peak-valley difference rate as high as 70%, thus classifying it as a peak-valley difference user. If a user is not a suitable peak-valley difference user for energy storage configuration, the process ends or redirects to other energy efficiency service recommendations; this is not within the scope of this solution. Step S2 focuses on basic electricity costs and parameter estimation, based on the time-of-use pricing policy in the user's location. π(t) Calculate the basic electricity cost for the current year C base Current year's basic electricity fee C base The calculation uses the dynamic integration method and the formula. Calculation, where for t Time-based load data, π( t )for t The time-of-use electricity price data for the user's location includes peak electricity price and valley electricity price. If the billing is not based on peak or valley electricity price, it is the average electricity price. The time-of-use electricity price data comes from the electricity price policy database of the data access and integration layer. In this embodiment, the time-of-use electricity price for the location of a certain manufacturing enterprise is RMB 1.2 / kWh for peak electricity and RMB 0.3 / kWh for valley electricity. Δt For time intervals, T This represents the total number of time intervals after being divided according to a set time interval. This method encourages users to utilize high loads during off-peak hours, thereby contributing to peak shaving and valley filling for the power grid. In this embodiment, the annual electricity cost for the manufacturing enterprise is calculated to be approximately 5 million yuan. Of course, other methods can also be used to calculate the basic electricity cost, such as the static extreme value method, which uses the maximum instantaneous power consumption to pay a fixed fee. This incentivizes users to actively reduce their maximum load by managing production processes and configuring energy storage. This step also obtains the unit capacity cost C of the energy storage system. cap Charge / discharge efficiency η, cycle life N cycle Local energy storage discharge subsidy standard S sub Energy storage technology parameters, S sub The unit is yuan / kWh, and these parameters come from the equipment parameter library of the data access and integration layer. The automatic peak-valley difference user identification algorithm based on load curves realizes the transformation from manual experience-based judgment to data-driven automatic identification, with an identification accuracy of ≥95%. The identification time is shortened from the traditional 2-3 days to within 5 minutes, and the data processing scale can support simultaneous analysis of more than 10,000 enterprise users.

[0016] S3 steps focus on maximum configurable capacity E max Confirmed, considering the installation space constraints of the users. Transformer capacity constraints and investment budget constraints Determine the configurable upper limit of the energy storage system's capacity. ,in Constraint calculation uses , , . The upper limit of subsequent traversal calculations directly determines the search space, ensuring that all evaluated capacity solutions are physically feasible, electrically safe, and economically affordable. This embodiment defines the maximum configurable capacity for the manufacturing enterprises it connects to. The capacity is 2000 kWh. S4 focuses on the economically optimal capacity configuration traversal calculation with a preset step size. ΔE In this embodiment, as an increment, ΔE Choose 1kWh, from a capacity of 0 to... E max Iterate through the values, for each candidate capacity E i Calculate its initial investment cost I initial ( E i The system simulates its operation over its entire lifespan of N years and calculates the benefits including electricity cost savings. B saving (y) Annual net income from operating and maintenance costs B net (y,E i ) Then calculate the net present value of the capacity plan. NPV (E i ) The calculation of the economically optimal capacity allocation specifically includes: using the formula for initial investment cost. Calculation, where Fixed costs that are independent of capacity.

[0017] The net present value The calculation formula is: ,in r The discount rate is... N For the entire period of financial analysis, the number of years is [number]. N Typically linked to the lifespan of energy storage devices, but not their physical lifespan; rather, it represents the payback period for investing in and deploying energy storage; the annual net return on operating and maintenance costs. The calculation formula is Among them, the benefits of electricity cost savings The calculation formula is: ,in for y year t Discharge power at any given time Δt For time intervals, for tPeak electricity price at any given time for t Off-peak electricity prices at specific times; for y The annual battery replacement cost is calculated based on the battery reaching the end of its lifespan. In this embodiment, when a manufacturing enterprise is configured with a simulated 500kWh capacity, the system simulates charging during off-peak hours and discharging during peak hours based on electricity prices and load curves, calculating an annual electricity cost saving of approximately 250,000 yuan.

[0018] When considering energy storage subsidies, the annual net benefit of operation and maintenance costs Use formula ,in for y The annual electricity subsidy income is calculated using the following formula: ,in for y Total annual discharge S sub The amount of subsidy for the unit in the user's location. , Based on P load (t) , π(t)、 The system uses η and preset, unoptimized charge / discharge strategy rules (such as peak-to-valley charging and peak-to-peak discharging) that are independent of specific capacity to simulate the daily operation of an energy storage system over its entire lifespan of N years. The calculated annual charge / discharge volume can also be used to assess the health of the energy storage system's batteries. The charge / discharge strategy rules here are relatively simple. For example, during off-peak hours, if the energy storage battery has available capacity, it is charged at its maximum permissible power. During peak hours, if the energy storage battery has power and there is load demand, it is discharged at its maximum permissible power, independent of capacity. E i The charge and discharge power is stored in the energy storage system. t The calculation of this intermediate value of charging and discharging power at any given moment depends on the load power. P load (t)、 Time-of-use electricity pricing π(t)、 Charge / discharge efficiency η 、 Real-time battery status. This step also allows for the calculation of the payback period (PBP): PBP = Initial investment / Electricity cost savings.

[0019] Compared to traditional experience-based trial-and-error methods or search methods using large step sizes, this algorithm sacrifices a small amount of computation time—typically within 30 seconds—to achieve an absolute guarantee of accuracy and optimality in the configuration results. Compared to traditional locally optimal configuration methods, the return on investment (ROI) of energy storage configuration can be increased from 8%–12% to 15–25%, an improvement of 50–100%.

[0020] Step S5 focuses on determining the optimal solution and generating the charging / discharging strategy, reflecting a refined economic assessment based on a full life-cycle cash flow model. For each capacity to be evaluated, the algorithm performs a complete time-series-based cash flow simulation: comparing all candidate capacities. E i corresponding NPV(E i ) Select to make NPV(E i ) Maximum capacity E opt As the optimal economic allocation solution, and based on E opt , P load (t) and π(t) Generate a charge / discharge strategy plan table with specific time granularity. Schedule(t) During this calculation process, settings can also be configured within the loop body. The system uses a decision node to dynamically update the optimal solution, employing global search and real-time optimization. The generation method can utilize solvers such as linear programming solvers or mixed-integer programming solvers to solve for the optimization objective of minimizing the daily total electricity cost under constraints. Solver tools include PuLP, CBC, Gurobi, and CPLEX. In this embodiment, after reviewing all options, the manufacturing enterprise found that configuring 1200kWh resulted in the highest net present value over the entire project lifecycle, reaching 1.8 million yuan. That is, 1200kWh, and the system then generates the optimal charge and discharge schedule for 1200kWh. It is the globally optimal solution that maximizes the Net Present Value (NPV), rather than an empirically determined local optimum. The tight coupling design between the charge / discharge power generation in S5 and the charge / discharge power simulation in S4 ensures that the economically optimal capacity is necessarily matched with an executable operating strategy validated in the simulation, avoiding the technical risk of capacity and strategy becoming disconnected. The dynamic charge / discharge strategy optimization engine, based on real-time electricity prices and load forecasting, implements an adaptive charge / discharge strategy, upgrading from a fixed strategy to dynamic optimization. Compared to a fixed strategy, it increases electricity cost savings by an additional 8-12% and extends battery life by 15-20% by optimizing charge / discharge depth. The generated charge / discharge strategy schedule is shown. Schedule(t) This includes a sequence of charging and discharging power commands, which are converted into control commands and sent to the battery management system (BMS) to guide the energy storage system in performing automated peak shaving and valley filling operations. The charging and discharging strategy plan for a specific day generated by a manufacturing enterprise in this embodiment is as follows:

[0021] S6 focuses on the automatic generation and report output of ROI (Return on Investment) models, based on E... opt The system automatically generates an investment analysis report, including but not limited to the Payback Period (PBP), Internal Rate of Return (IRR), and visual charts, based on the corresponding cash flow forecast data. The IRR, which satisfies NPV=0, considers the time value of money and more scientifically reflects the true profitability of energy storage than the simple Return on Investment (ROI). In this example, the initial investment for the manufacturing enterprise is approximately 1.5 million yuan, with annual electricity cost savings of approximately 600,000 yuan, a Payback Period (PBP) of approximately 2.5 years, and an IRR greater than 20%. This report is directly pushed to the customer through the system. Visual charts include, but are not limited to, a full lifecycle cash flow diagram and a revenue-capacity relationship curve.

[0022] The time to generate a full life cycle investment return model report is reduced from 2-3 days to within 3 minutes, the investment payback period prediction error is less than ±0.3 years, investment risk is quantified, and the success rate is increased to 90%.

[0023] Based on the same inventive concept, this application also implements it through the following embodiments. These embodiments are not simply a stacking of software functions, but rather a collaborative computing platform with data-driven core, algorithmic models as the engine, and deeply integrated with the aforementioned methods. This reflects a clear technological orientation, aiming to specifically solve the problems of "lack of analysis tools, decision-making relying on experience, and fragmented service processes" mentioned in the background art through systematic software and hardware coupling. Based on a decoupled architecture driven by a dual-core "analysis engine-service platform," core, computationally intensive algorithmic models such as load feature extractors and economic optimal configuration solvers are encapsulated into an independent "core analysis engine," while business logic, process management, and user interaction functions are built on a "full-lifecycle service management platform." The two exchange data and schedule tasks through standardized APIs. This allows the economic optimal capacity traversal algorithm to run efficiently and stably without interruption by business flows, improving the system's computational performance and reliability. The analysis engine can be upgraded independently, such as to create a more accurate battery degradation model, without reconstructing the entire business platform, improving the system's maintainability and technological evolution capabilities. The service platform can trigger the core engine with one click to complete the full analysis chain from step S1 to step S6, ensuring that the comprehensive online and offline full-cycle service method for major customers' energy is realized in the system as a coherent and automated technical process.

[0024] A comprehensive online and offline full-cycle energy service system for large customers, used to implement the aforementioned method, includes: a data access and integration layer, a core analysis engine, a full-cycle service management platform, and a multi-terminal interaction and control system. The data access and integration layer is used to connect with the user's smart meter or energy management system (EMS) via a standardized communication protocol to collect load time-series data. Pload ( t And store the time-of-use electricity pricing model. π(t) The data access and integration layer connects to the user-side smart meters or energy management systems (EMS), and includes parameters for energy storage devices. The standardized communication protocol can be the DL / T 645 multi-functional energy meter communication protocol or the Modbus TCP protocol. The data access and integration layer connects to the power grid data system, including the user-side smart meters or energy management system (EMS). The power grid data system, along with enterprise ERP systems, equipment monitoring platforms, and policy information sources, are all part of various external systems, and these external systems are not limited to these types. In this embodiment, the data access and integration layer includes an electricity data access library, an electricity price policy library, and an equipment parameter library. The power grid data system can automatically collect and upload time-series load data at a preset frequency of 15 minutes, consistent with the aforementioned time intervals. P load ( t ) to the electricity data access database ， Time-of-use electricity pricing model π(t) Information is collected from policy sources and stored in the electricity price policy database. Energy storage equipment parameters are stored in the equipment parameter database, including capacity-based electricity prices and the unit capacity cost (C) of the energy storage equipment. cap Charge / discharge efficiency η, cycle life N cycle These databases, used for subsequent calculations, are all structured databases that support version management and regional parameter matching. Data cleaning, normalization, and storage are also required before data processing.

[0025] The core analysis engine is connected to the data access and integration layer and is used to perform calculations such as load feature extraction, economic optimal configuration solving, and charging / discharging strategy generation. The core analysis engine is the technical hub of the system, hosted by multiple computing servers, and internally encapsulates the core algorithms for implementing the aforementioned comprehensive online and offline full-cycle energy service method for large customers. In this embodiment, the core analysis engine includes a load feature extractor, an economic optimal configuration solver, and a charging / discharging strategy generator. The load feature extractor is used to process load time-series data... P load (t) Process it and call the peak-valley difference rate. R peak-valley The eigenvalue calculation algorithm outputs a structured user load feature vector; the economic optimal configuration solver incorporates a traversal algorithm, receiving feature vectors from the load feature extractor, electricity price parameters from the electricity price policy database, and the maximum configurable capacity of the energy storage system. E max Used to achieve high-precision step sizes within the constraints of physical capacity limits. ΔE Iterate through and calculate the capacity of all candidates in a loop. E i Net present value NPV(E i) and output the globally optimal capacity. E opt and the corresponding expected return curve; the charging / discharging strategy generator is a strategy compilation unit used to base on E opt Real-time or predictive P load (t) and π(t) Applications such as cost minimization optimization rules generate time-indexed charge / discharge strategy plans for specific future periods, such as 24-hour periods. Schedule(t) The charging and discharging strategy plan table Schedule(t) It contains a sequence of charging and discharging power commands, which can be directly parsed and executed by downstream control systems.

[0026] Since the algorithm's input parameters are all stored as externally configurable variables in the electricity price policy database and equipment parameter database, when these parameters change, only the database needs to be updated and the algorithm rerun to obtain the new values; the algorithm itself does not need to be modified. After the main algorithm flow is complete, the system can also automatically execute the following single-factor analysis and scenario analysis sensitivity analysis routines: Single-factor analysis focuses on key parameters such as electricity price difference and investment cost, recalculating E based on variations within the positive and negative range of the benchmark value. opt Combined with NPV, a Tornado Chart is generated to quantify the impact of each parameter on returns; scenario analysis is preset with multiple future scenarios such as a 10% increase in electricity prices and subsidy reduction, and the algorithm is run in batches to output the optimal configuration and return range under different scenarios, providing a risk spectrum for investment decisions.

[0027] The full-cycle service management platform and the core analysis engine form a collaborative working environment driven by data models and rule engines. It is tightly integrated with the core analysis engine through API interfaces to manage customer technical files, advance business opportunity processes, and automatically generate structured investment analysis reports. The full-cycle service management platform includes a customer and technology file management module, an intelligent business opportunity promotion pipeline module, a structured report generation and rendering service module, and a knowledge graph and case reasoning library. The entire system does not store isolated customer information or analysis reports. Instead, it constructs a unified data entity centered on a "customer technical profile." This entity dynamically links original workload data, characteristic parameters, previous analysis results, generated strategies, and final project data, thereby achieving a unified data governance architecture oriented towards the "full-cycle data entity." This not only forms a closed-loop data verification and optimization foundation from customer information to potential business opportunities to technical and economic analysis to agreement signing to case accumulation, but also achieves traceability of technical analysis. Any output investment analysis report can be traced back to the original data and calculation process it was based on, enhancing the scientific rigor and credibility of the solutions.

[0028] The customer and technical profile management module creates a unique digital profile for each customer. This profile not only stores basic information but also dynamically links to their load feature vectors, such as... E op Historical analysis records of t and NPV, along with equipment operation data, form a queryable and analyzable data entity. In this embodiment, the customer and technical file management module is a major customer information management module. This module is a dynamic customer technical characteristic database that continuously receives and integrates the output from the core analysis engine, constructing a database for each customer that includes information such as peak-valley difference rate. R peak−valley A profile of the system using technical tags such as historical load characteristics, results of previous optimization schemes, and final adopted configuration.

[0029] The intelligent business opportunity advancement pipeline module is configured as a state machine-driven process engine. Each business opportunity is modeled as a state object, used to automatically associate the changes in the business opportunity state with the analysis tasks of the core analysis engine. This allows for the automatic triggering of user load analysis or economically optimal capacity configuration traversal calculations during the business opportunity advancement process, such as from "intent" to "analysis" to "contract signing." In this embodiment, the intelligent business opportunity advancement pipeline module includes an intent and business opportunity management module, an integrated energy management module, and an integrated energy expert management module. The intent and business opportunity management module is used to mark potential customers or input preliminary load data. The integrated energy management module is used to automatically trigger or be manually initiated by technical personnel to complete a full-cycle online and offline integrated energy service method for large customers. Each analysis task is considered a technical operation, with its queueing, calculation, and completion statuses, and input data, such as... E opt The outputs and versions of NPV and Schedule(t) are all stored in the system and automatically associated with the business opportunity record to form a "business opportunity-technical solution" data pair, providing direct technical support for subsequent decision-making. The integrated energy expert management module is used to store expert domain tags, providing support for creating offline service tasks through the platform.

[0030] The structured report generation and rendering service module includes an ROI model estimation module, which is an automated report generation service module. Through a predefined API interface, it actively obtains the output of the core analysis engine after calculation, containing... E optThe system automatically compiles, fills, calculates, and renders structured data packages such as NPV curves and Schedule(t) based on pre-set financial models and document templates. It generates a visual investment analysis report containing detailed technical parameters, economic benefit charts, and sensitivity analysis, and formats it into a standard document output. The standard document can be an automatically synthesized standard report file containing technical solutions, financial analysis, and visual charts in PDF or PPT format.

[0031] The knowledge graph and case reasoning library are used to store anonymized successful project case data models. In this embodiment, it includes a successful case library, a comprehensive energy encyclopedia module, and a comprehensive energy expert management module. The successful case library stores historical cases and their configuration schemes, while the comprehensive energy encyclopedia module provides contextual support for the aforementioned technical analysis process. When analyzing new clients, during the analysis process, technicians or the system can perform real-time queries on specific technical points such as certain battery characteristics or interpretations of local subsidy policies. Based on load feature vector similarity, the system can automatically retrieve and recommend anonymized historical cases and their configuration schemes in the successful case library that have similar technical solutions and economic benefits. This provides analysts and clients with auxiliary data references and confidence level evidence for the analysis, enhancing the persuasiveness and technical credibility of the solutions. When the analysis encounters complex boundary conditions or requires on-site verification, the Comprehensive Energy Encyclopedia module, in conjunction with the Comprehensive Energy Expert Management module, can create offline service tasks through the platform. Based on the expert domain tags in the Comprehensive Energy Expert Management module, these tasks, such as on-site inspections and technical reviews, are precisely assigned to relevant internal and external experts. The experts' feedback is then used as structured information to populate the project archive, achieving seamless collaboration between online and offline technical capabilities. This embodiment's successful case library has accumulated over 120 validated industry solution templates, and the Comprehensive Energy Encyclopedia module has amassed over a thousand technical, policy, and market documents, becoming a valuable resource for internal corporate training and customer education.

[0032] The multi-terminal interaction and control system embodies the technology of online-offline integration. It consists of a set of front-end applications following the same back-end service protocol, providing access and interaction interfaces for different levels of functions to internal staff of offline service providers, on-site engineers, and enterprise users. For example, the multi-terminal interaction and control system pushes the visualized investment analysis report generated by the ROI model estimation module to customers in real time and records customer reviews and feedback, completing the automated conversion from technical results to decision-making documents. In this embodiment, the multi-terminal interaction and control system includes a responsive web console, a mobile on-site service terminal, and a customer-side mini-program / portal. The responsive web console is designed for internal experts and managers of service providers, providing a full-featured interface for parameter configuration, task scheduling, in-depth result analysis, and platform management of the core analysis engine using a PC-based web application. The mobile on-site service terminal is a mobile tool for marketing and engineering personnel, supporting both offline and online modes. It is used to collect and quickly import on-site load data via mobile web / app, and to call the lightweight computing service of the core analysis engine for solution simulation, realizing the on-site application of technical services. This tool simplifies and streamlines complex technical and economic analysis models, enabling marketing personnel to quickly generate in-depth preliminary solutions and significantly improving front-end response speed and sales conversion rates. The customer-side mini-program / Portal is a lightweight customer interface. In this embodiment, lightweight terminals such as WeChat mini-programs not only display information but also serve as entry points for customer-side data confirmation and channels for operational data feedback. Customers can authorize access to their electricity consumption data to view customized analysis reports and real-time operational status of the energy storage system, such as charging and discharging power and revenue accumulation, forming a service loop. The multi-terminal interaction and control system constructs a technical visualization and confirmation channel for the customer: customers can not only view the final report but also, after authorization, view load curves generated based on their own data and system recommendations. E opt This includes key intermediate technology results such as revenue forecasting. After project execution, the terminal can further connect to energy storage system monitoring data to present results to customers. Schedule ( t By tracking the actual implementation status and daily / monthly actual cost savings, the technical solutions, implementation effects, and customer perceptions are thoroughly integrated, forming a long-term service relationship based on data trust.

[0033] For the aforementioned manufacturing companies, from the moment a client is entered into the intention and opportunity management module, all their basic information, workload data, analysis reports, and signed agreements are recorded in the system. This project is automatically added to the success story database. Clients can view project progress and revenue reports through a mini-program.

[0034] To ensure the system's technical openness and scalability, the large customer integrated energy online and offline full-lifecycle service system of this application also includes a standardized API gateway. This API gateway provides a RESTful API interface, allowing third-party systems to securely access the system's data or trigger services. The system's data includes optimal strategies. Schedule ( t The third-party systems mentioned include enterprise ERP systems, power grid dispatching systems, or energy storage battery management systems (BMS). Triggering services include initiating a new economic analysis or issuing a sequence of charge / discharge power commands to the energy storage battery management system (BMS). A standardized API gateway ensures the universality and accuracy of the core algorithms, reduces the technical threshold and risk of system integration, and significantly enhances the system's industrial integration capabilities and practical value.

[0035] The system, through multi-terminal interaction and control systems and standardized API gateways, technically connects online analysis and computation with offline field services. The entire technical solution extends algorithmic capabilities to the service site, addressing the technical shortcomings of traditional models where on-site personnel cannot perform rapid quantitative analysis, thus improving service response speed. The charging and discharging strategies generated by the system are no longer static documents, but control commands that can directly drive physical devices through interfaces, truly realizing a closed-loop technology from "digital decision-making" to "physical execution." It also establishes a visualized trust channel on the customer side, solidifying data-driven technical service relationships.

[0036] This solution was applied to another auto parts manufacturing plant, which typically has a daily load peak-valley difference rate of 65% and an annual electricity cost of approximately 12 million yuan. Through traversal algorithms, the economically optimal energy storage configuration was determined to be 1.2 MWh, with an initial investment of approximately 1.8 million yuan, annual electricity cost savings of approximately 350,000 yuan, and the investment payback period was increased from the traditional experience-based approach of 6.5 years to approximately 4.1 years. The net benefit over the entire project lifecycle is approximately 3.2 million yuan.

[0037] Through the comprehensive online and offline full-cycle energy service system for major customers, project implementation cycles are shortened by 40-60%; personnel efficiency increases by 3-5 times the number of customers served per person, thereby reducing service costs by 50-70%; investment decision accuracy improves from an error rate of 25-40% using traditional experience-based methods to an error rate of <5%, an improvement of over 80%; electricity cost reduction increases from 10-15% with a single strategy to 18-27% with dynamic optimization, an improvement of 60-80%; the payback period (PBP) is reduced from the traditional 5-8 years to 3.5-5.5 years, a reduction of 30-40%; solution generation time is improved from the traditional 3-5 working days to within 30 seconds, an efficiency improvement of 99%; configuration accuracy is improved from the 50-100kWh level to the 1kWh level, an improvement of 50-100 times; and the solution adoption rate increases from 45% using traditional methods to 80%. 5.7%; Data processing capabilities have increased from single-user analysis to concurrent analysis of tens of thousands of users, a thousand-fold increase in scale; Prediction accuracy has improved from an empirical estimation error of 15-25% to an error of <5%, a 3-5 fold increase in accuracy; System availability has expanded from a traditional single PC terminal to 99.5% availability with multi-terminal coverage, significantly broadening the service scope; Over-investment risk has decreased from a loss rate of 20-30% to 80-90%; Under-configuration risk has decreased from an opportunity loss of 15-25% to 85-95%; Technology selection risk has decreased by 70-80%; The probability of underperformance has decreased from 30-40% to a deviation of <5%, a reduction of 85-90%; Service interruption risk has decreased by over 95%; Energy utilization efficiency has improved grid efficiency by 15-20%; Grid investment has been reduced by 2-3 million yuan per MW; Carbon emissions have been reduced by 0.8-1.2 tons per MWh.

[0038] The above description is only a preferred embodiment of the present solution, but the scope of protection claimed by the present solution is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in this application, based on the technical solution and inventive concept of this application, should be included within the scope of protection of this application.

Claims

1. A comprehensive online and offline full-cycle energy service method for large customers, characterized in that, Includes the following steps: S1 user load analysis and type determination: Obtain electricity consumption data from enterprise users and generate time-series load data. P load ( t The typical daily load power curve of ) t Calculate the peak-to-valley difference rate of the curve at the given time of day. R peak-valley Using formula Calculation, where Maximum daily load Minimum daily load, and compared with a preset threshold. If a comparison is made, If the peak-valley difference is not specified, the user is determined to be a suitable user for configuring energy storage; S2 calculates the current year's basic electricity fee. Using formula Calculation, where for t Time-based load data, π( t )for t Time-of-use electricity price data for the user's location at any given time. Δt For time intervals, T This refers to the total number of time intervals after dividing the data according to a set time interval, and includes the unit capacity cost C. cap Charge / discharge efficiency η, cycle life N cycle S3 determines the energy storage system parameters; S3 determines the maximum configurable capacity of the energy storage system. E max , ,in Install space constraints for users. Due to user transformer capacity constraints, Constraining user investment budgets; S4 uses a preset step size ΔE As an increment, from the capacity value of 0 to... E max Iterate through the values, for i For each candidate capacity of the index number E i Calculate its initial investment cost I initial (E i ) It simulates its operation over its entire lifespan of N years and calculates the benefits including electricity cost savings. B saving (y) Annual net income from operating and maintenance costs B net (y, E i ) Then calculate the net present value of the capacity plan. NPV(E i ) ,in y For the year; S5 compares all candidate capacities. E i Corresponding net present value NPV(E i ) Select the option that makes the net present value (NPV) NPV (E i ) Maximum capacity E opt As the optimal economic allocation solution, and based on E opt , P load (t) and π(t) Generate a time-granular charge / discharge strategy plan table containing a sequence of charge / discharge power commands. Schedule(t) The charging and discharging power command sequence is used to guide the energy storage system to perform automated peak shaving and valley filling operations; S6 is based on E opt It automatically generates an investment analysis report containing the payback period (PBP), internal rate of return (IRR), and visual charts, based on the corresponding cash flow forecast data.

2. The method for providing comprehensive online and offline full-cycle energy services to large customers according to claim 1, characterized in that, Step S4, the calculation of the economically optimal capacity configuration, specifically includes: the initial investment cost. The calculation formula is: ,in The net present value is a fixed cost independent of capacity. The calculation formula is: ,in r The discount rate is... N For the full lifecycle of financial analysis; annual net revenue from operating costs. The calculation formula is ,in for y The annual electricity cost savings are calculated using the following formula: ,in for y year t Discharge power at any given time Δt For time intervals, π peak ( t )for t Peak electricity price at any given time π valley ( t )for t Off-peak electricity prices at specific times; for y Annual battery replacement cost.

3. The method for providing comprehensive online and offline full-cycle energy services to large customers according to claim 2, characterized in that: Annual net income from operation and maintenance costs Use formula ,in for y The annual electricity subsidy income is calculated using the following formula: ,in for y Total annual discharge Ssub The amount of subsidy is given to the unit in the user's location.

4. A comprehensive online and offline full-cycle energy service system for large customers, characterized in that: It includes a data access and integration layer, a core analysis engine, a full-lifecycle service management platform, and a multi-terminal interaction and control system. The data access and integration layer is used to collect load time-series data through standardized communication protocols. P load (t) And store the time-of-use electricity pricing model. π(t) The system includes energy storage system parameters; the core analysis engine is connected to the data access and integration layer, and is used to perform calculations for load feature extraction, economic optimal configuration solution, and charging and discharging strategy generation; the full-cycle service management platform is integrated with the core analysis engine through an API interface, and is used to manage customer technical files, advance business opportunity processes, and automatically generate structured investment analysis reports. The business opportunity advancement is used to trigger the implementation of the large customer integrated energy online and offline full-cycle service method and the creation of offline service tasks as described in any one of claims 1 to 3; the multi-terminal interaction and control system is used to provide access and interaction interfaces for different levels of functions for offline service provider internal personnel, on-site engineering personnel, and enterprise users.

5. The large customer integrated energy online and offline full-cycle service system according to claim 4, characterized in that: The core analysis engine includes a load feature extractor, an economic optimal configuration solver, and a charging / discharging strategy generator; the load feature extractor is used to process load time-series data. P load (t) Process the data and calculate the peak-to-valley difference rate. R peak-valley The economic optimal allocation solver incorporates a traversal algorithm to address capacity limitations. E max Within the constraints, with step size ΔE Traverse and calculate all candidate capacities E i Net present value NPV(E i ) and output the globally optimal capacity. E opt The charge / discharge strategy generator is used to base on... E opt , P load (t) and π(t) Generate charge / discharge strategy plan table Schedule(t).

6. The large customer integrated energy online and offline full-cycle service system according to claim 4, characterized in that: The full-cycle service management platform includes an intelligent opportunity promotion pipeline module and a structured report generation and rendering service module. The intelligent opportunity promotion pipeline is configured as a business opportunity state machine-driven process engine, used to automatically associate changes in business opportunity states with the analysis tasks of the core analysis engine, enabling automatic triggering of user load analysis or economically optimal capacity configuration traversal calculations during opportunity promotion. The structured report generation and rendering service is used to call preset document templates and generate reports from the core analysis engine. E opt , NPV and Schedule(t) The data is automatically populated and formatted for standard document output.

7. The large customer integrated energy online and offline full-cycle service system according to claim 4, characterized in that: The multi-terminal interaction and control system includes a mobile field service terminal, which supports offline and online modes. It is used to quickly import field load data and call the lightweight computing service of the core analysis engine to perform scheme simulation.

8. The large customer integrated energy online and offline full-cycle service system according to claim 4, characterized in that: It also includes system integration and extension interfaces, which provide standardized RESTful API interfaces for third-party systems to access data or trigger services. These third-party systems include enterprise ERP systems, power grid dispatching systems, or energy storage battery management systems (BMS).

9. The large customer integrated energy online and offline full-cycle service system according to claim 4, characterized in that: The full-cycle service management platform also includes a knowledge graph and case reasoning library, which stores anonymized historical project case data models. When analyzing new clients, it automatically retrieves and recommends similar historical cases and their configuration schemes based on load feature similarity.

10. The large customer integrated energy online and offline full-cycle service system according to claim 4, characterized in that: The multi-terminal interaction and control system includes a client-side mini-program or portal for enterprise users. This interface is used not only to display analysis reports and revenue forecasts to customers, but also to serve as an entry point for customer-side data confirmation and a feedback channel for real-time operation data of the energy storage system.

Images