Water plant station light storage construction project investment intelligent evaluation method and system

Abstract

The invention provides a water plant station light storage construction project investment intelligent evaluation method and system, and relates to the technical field of industrial plant station energy management and investment evaluation, the system comprises a hardware layer and a function layer, and the components of the hardware layer adopt an industrial standard communication protocol to realize data interaction and instruction transmission; the hardware layer comprises a sensor network, a monitoring center, a comprehensive management platform, an algorithm server, an equipment regulation and control center and a central database; the monitoring center is in communication connection with the sensor network and is used for carrying out denoising and missing value complementation preprocessing on the acquired data; a plurality of typical daily load curves are accurately extracted from a high-dimensional noisy water affair load time sequence, and hidden and dispersed load clustering modes are effectively recognized; the matching relation between each typical daily load and the corresponding illumination characteristic is microscopically analyzed, and the overall characteristic and the change rule of the annual load are macroscopically integrated, so that the photovoltaic capacity configuration and the energy storage unit type selection can be ensured to be deeply adaptive to the actual energy demand of the water plant station.

Description

Technical Field

[0001] This invention relates to the field of energy management and investment assessment technology for industrial power plants, specifically to a smart assessment method and system for investment in photovoltaic and energy storage projects for water plants. Background Technology

[0002] In recent years, with the increasing awareness of environmental protection and the advancement of carbon peaking and carbon neutrality strategies, the photovoltaic and energy storage industries have developed rapidly. Waterworks, sewage treatment plants, and other high-energy-consuming industries have found it crucial to construct photovoltaic + energy storage systems as an important way to reduce reliance on traditional electricity, optimize electricity costs, and improve overall energy efficiency. Waterworks possess abundant photovoltaic installation resources, such as rooftops, above water tanks, and unused land. The construction of photovoltaic-energy storage systems can fully utilize these resources, achieving solar energy conversion and surplus energy storage, enhancing the flexibility and reliability of energy dispatch, and meeting the industry's energy conservation and emission reduction development needs. However, existing photovoltaic and energy storage project assessment and management systems have significant shortcomings. They are difficult to adapt to the specific needs of water plants and ignore the load characteristics of water plants. The power load of water plants is high-dimensional time series data, containing a lot of noise and exhibiting complex patterns at multiple time scales, such as daily, weekly, seasonal fluctuations and holiday effects. The differences in different water supply load patterns lead to extremely irregular shapes, densities, and distributions of load clusters. Traditional clustering methods are difficult to accurately characterize their internal structure. Existing analysis schemes are mostly limited to macro-level total analysis or isolated typical day assessments, which easily mask daytime fluctuations, extreme load phenomena, or ignore annual seasonal changes and energy balance. This results in a disconnect between photovoltaic and energy storage configuration and actual energy consumption scenarios, affecting the rationality and economy of system design. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides an intelligent investment assessment system for water plant photovoltaic and energy storage projects, comprising a hardware layer and a functional layer. The components of the hardware layer use industrial standard communication protocols to achieve data interaction and command transmission. It should be noted that the specific protocols include Modbus, OPCUA, or EtherCAT, which are commonly used protocols in water plant energy management scenarios, taking into account both compatibility and transmission efficiency. The hardware layer includes sensor networks, monitoring centers, integrated management platforms, algorithm servers, equipment control centers, and central databases. Sensor networks are deployed at water plant load nodes, photovoltaic module arrays, and energy storage battery packs to collect real-time power load data, photovoltaic output data, and energy storage system operating status data. The monitoring center communicates with the sensor network to perform noise reduction and missing value completion preprocessing on the collected data. The integrated management platform communicates bidirectionally with the monitoring center, algorithm server, equipment control center, and central database. It serves as the core hub for system data flow and control command issuance. The central database is used to store historical power load / photovoltaic output / energy storage status data collected by the sensor network, calculation and evaluation results output by the algorithm server, and expert experience database data. It also supports data retrieval from the integrated management platform and result backtracking from the algorithm server. The calculation and evaluation module is implemented by the integrated management platform and the algorithm server. It is used to perform load characteristic analysis, photovoltaic and storage system optimization configuration, project cost estimation, full life cycle economic evaluation and post-commissioning evaluation of water plant photovoltaic and storage projects. The intelligent management module is implemented independently by the integrated management platform and is used to realize the visualization of key project data, support operational decision-making, and control system access security.

[0004] Preferably, the calculation and evaluation module includes the following sub-modules: The basic information integration submodule is used to input and store system basic parameters, financial basic parameters, and operational basic parameters; The system's basic parameters include photovoltaic module efficiency, energy storage battery type, building installation area, and module size; Financial parameters include benchmark rate of return, loan interest rate, and tax rate; operational parameters include annual equipment degradation rate, annual maintenance cost, and energy storage battery cycle life. The load characteristic analysis submodule is used to extract representative daily load curves from the high-dimensional noisy power load time series data of the water plant using the K-medoids clustering algorithm based on Levy flight; the photovoltaic optimization configuration submodule is used to calculate the maximum installable capacity and recommended configuration capacity of photovoltaic modules by combining the priority of the building's installable area, local solar resource data and the water plant's load absorption capacity. The priority of areas where buildings can be installed is divided into the highest priority and the second highest priority. The energy storage optimization configuration submodule is used to determine the rated power, rated capacity, and charging and discharging strategy of the energy storage system based on the photovoltaic output curve, water plant load characteristics, peak-valley electricity pricing policy, and inverter total capacity constraints, and to perform physical layout feasibility verification; the cost estimation submodule is used to estimate the equipment purchase and installation costs, construction costs, and other construction costs according to the photovoltaic and energy storage system configuration results, and to form the total investment estimate of the project. The economic evaluation submodule is used to build dynamic financial models and calculate the project's internal rate of return, payback period, and total return on investment. The post-evaluation submodule is used to compare the actual operating data after the project is put into operation with the preliminary calculation results, and output targeted optimization suggestions.

[0005] Preferably, when the load characteristic analysis submodule executes the K-medoids clustering algorithm based on the Levy flight improvement, it includes the following steps: Step 201: Randomly select K initial Medoids, where K ranges from 3 to 8, calculate the Manhattan distance from each load data point to the initial Medoids, and determine the initial clusters and the total clustering cost; Step 202: Assign all points to the nearest Medoids; The formula for calculating Manhattan distance is: ; Where xᵢ is the i-th load data point, m j Let t be the j-th Medoid, and t be the time dimension; For each data point xᵢ, calculate its path to all Medoids. The distance to the Medoid is calculated, and the current total cost is the sum of the distances between all sample points and the Medoid. The specific formula is as follows: ; Step 203: Iterate through all non-Medoid sample points, and try to replace each current Medoid with a non-Medoid sample point in turn, calculating the cost change after the replacement. ,in The total cost after replacement. Total cost before replacement; Step 204: If ΔCost < 0, perform the optimal replacement, which is to select the scheme with the smallest ΔCost; if ΔCost ≥ 0, generate a random number in the interval between 0 and 1. If the random number is less than the preset trigger probability p, randomly select 1 from the current Medoids as the perturbation reference point.

[0006] It should be noted that the preset trigger probability p is determined based on the noise intensity of the water plant load data and the number of clusters K: when the load data noise intensity is >15% (the noise intensity is calculated based on the standard deviation / mean of the data), p is 0.2-0.3; when the load data noise intensity is ≤15%, p is 0.1-0.2; at the same time, when the number of clusters K=3-5, p is 0.15-0.3, and when K=6-8, p is 0.1-0.25, ensuring that the algorithm can avoid local optima under different data quality and clustering requirements, and the overall value of p is in the range of 0.1 to 0.3.

[0007] Preferably, when the load characteristic analysis submodule executes the K-medoids clustering algorithm based on Levy flight improvement, determining whether to trigger Levy flight includes the following steps: Step 205: To avoid the algorithm getting stuck in a local optimum, set the Levy flight trigger probability p and generate a random number between (0,1). If the random number is less than p, it can be considered that the Levy flight has been triggered and proceed to step 206; otherwise, proceed to step 209. Step 206: Select the disturbance center; Randomly select a Medoid from the current Medoids set M. m p As a reference point for disturbance; Step 207: Levy flight generates new candidate points; First, based on the Levy distribution, random Levy numbers need to be generated, and these will be used as the step size for Levy flight: ; ; ; ; Where u and v are normally distributed random variables; β1 is the characteristic index of the Levy distribution; Let v be the standard deviation of v; SL be the Levy flight stride. It is a gamma function; Map the continuous step size to discrete data point indices, and calculate the distance from all points to m. p Sort the distances and construct the distance-sorted sequence R: ; Where n is the total number of sample points in the dataset; Then, the ranking position r based on the Levy step size is calculated using the following formula: ; The final new Medoid index is determined as follows: ; in New candidate points selected based on Levy flight; Step 208: Force update the Medoid collection; Use new candidate points replace m p And update the Medoids set M to complete this perturbation update; Step 209: If the Medoids set changes, return to step 202 and start the next iteration; otherwise, end the algorithm and output the final result, which includes K clustering results. Finally, the Medoids set M and the representative load curves for each cluster are obtained.

[0008] Preferred: In the photovoltaic optimization configuration submodule, the highest priority for the building installation area is the space above the pool, the idle open space in the factory, and the unobstructed concrete roof; the next priority is the partially obstructed concrete roof and the corrugated steel roof. The formula for calculating the maximum number of photovoltaic modules that can be installed is: ; in Let be the effective area of ​​the installation area for the i-th building. is the photovoltaic module layout coefficient for the i-th installation area; in this embodiment, the value is 0.85-0.95, determined based on the flatness and shading conditions of the area; L PV The length of the photovoltaic module, l PV The width of photovoltaic modules, Ag PV Photovoltaic module installation azimuth angle, β Photovoltaic module installation tilt angle α The solar altitude angle of the area where the water plant is located. w PV The width of the photovoltaic module; The theoretical maximum total installed capacity PPVmax is calculated using the following formula: ; in PV,p This refers to the power of a single photovoltaic module.

[0009] Preferred configuration: The photovoltaic module has an initial capacity of P PV,RC The final calculation formula is as follows: ; Take immediately and The smaller value in the range.

[0010] Preferably, in the energy storage optimization configuration submodule, the rated discharge power of the energy storage system meets the following constraints: ; in The rated discharge power of the energy storage system. P PV Recommended capacity configuration for photovoltaic systems. The total capacity of the inverter at the point of common coupling; The rated capacity of the energy storage system is determined through a time-series simulation model.

[0011] Preferred: The intelligent management module includes the following sub-modules: The project statistics dashboard module is used to visually display basic project information, investment cost details, key operational indicators and economic evaluation indicators through bar charts, line trend charts, data tables and ranking lists. It supports filtering data by daily, weekly, monthly and yearly time dimensions. Basic project information includes construction scale and construction period; investment cost details include equipment cost, construction cost, and other cost percentages; key operational indicators include photovoltaic self-consumption rate and energy storage charging and discharging efficiency; economic evaluation indicators include internal rate of return and investment payback period. The cockpit submodule is used to integrate an electronic map engine to mark the geographical coordinates of each water plant. The map displays the real-time power generation of each water plant by default using color gradients. The color gradient division standard is red > 80% of rated power, yellow 40% to 80% of rated power, and green < 40% of rated power. Clicking on a single water plant icon triggers a floating data card that displays the plant's real-time power generation, real-time power consumption, power surplus or deficit, photovoltaic self-consumption rate, and remaining energy storage capacity or charging / discharging progress. It also provides a quick access point to the detailed power generation and power consumption statistics analysis interface for that water plant. The user management submodule is used to input, edit, and query user information, as well as disable or enable accounts. It supports user password reset and records user login logs. User information includes username, department, contact number, and job title. User login logs include login time, login IP, and login status. The system management submodule is used to configure a multi-level tree-structured department architecture, define role permissions, assign roles to users, and record user operation logs. It supports searching operation logs by keywords. The multi-level tree-structured department architecture supports department merging and splitting operations. Role permissions are associated with system menus, operation buttons, and the viewing scope of monitoring point data. User operation logs include operation time, operator, operation content, and operation result.

[0012] Preferred method: Also includes a smart assessment method for investment in water plant photovoltaic storage projects, comprising the following steps: S1: Input system basic parameters, financial basic parameters, and operational basic parameters through the basic information integration submodule and store them in the central database to provide data support for subsequent analysis. Among them, the system basic parameters include at least photovoltaic module efficiency, energy storage battery type, building installation area and module size; the financial basic parameters include at least benchmark rate of return, loan interest rate and tax rate; and the operational basic parameters include at least equipment annual degradation rate, annual operation and maintenance cost and energy storage battery cycle life. S2: The load characteristic analysis submodule calls the algorithm server to execute the K-medoids clustering algorithm based on Levy flight improvement. Input the water plant's historical power load data for the past year or more, and output K typical daily load curves and the annual frequency of each curve; the historical power load data must cover the energy consumption characteristics of the entire year. S3: Import building installation area survey data through the photovoltaic optimization configuration submodule, determine area priority, calculate the maximum installable capacity PPVmax and recommended configuration capacity PPVRC of photovoltaic modules, and generate a photovoltaic system configuration scheme; the building installation area survey data includes the area of ​​each area, shading status, and load-bearing capacity; the photovoltaic system configuration scheme includes module selection, installation area distribution, and array arrangement diagram. S4: Import the typical daily load curve output from step S2, the photovoltaic output prediction curve output from step S3, and the local time-of-use electricity price data through the energy storage optimization configuration submodule. Determine the preliminary electrical parameters of the energy storage system based on inverter capacity constraints and time-series simulation models. Combine the parameters with physical layout feasibility verification and adjustment to finally generate the energy storage system configuration scheme. The preliminary electrical parameters of the energy storage system include rated power PESdcmax and rated capacity EES. The energy storage system configuration scheme includes battery type, charging and discharging strategy, installation location planning, and fire protection design requirements. S5: Based on the configuration schemes in steps S3 and S4, the cost estimation submodule estimates the equipment purchase and installation costs, construction costs, and other construction costs by category, adds contingency funds to form the total project investment, and outputs the total investment estimate table. S6: Construct a dynamic financial model through the economic evaluation submodule. Input the total investment in step S5, the annual photovoltaic power generation revenue in step S3, the energy storage peak-valley arbitrage revenue in step S4, the annual operation and maintenance costs and tax rates in the basic operation parameters, calculate the annual cash flow according to the calculation period of conventional projects in the industry, and output the internal rate of return on investment, investment payback period, total investment return rate and net present value. S7: After the project is put into operation, the actual operation data is collected through the post-evaluation submodule and compared with the calculation results in step S6. The actual operation data includes monthly power generation, load factor, energy storage charging and discharging efficiency, operation and maintenance costs, and carbon emissions. Optimization suggestions include photovoltaic module tilt angle adjustment, energy storage charging and discharging time period optimization, and operation and maintenance cycle adjustment. S8: The project statistics dashboard of the intelligent management module and the cockpit sub-module display key data of steps S2-S7 in real time. Access control is achieved through user management and system management sub-modules to ensure the safe operation of the system and the compliant use of data.

[0013] It should be noted that the construction of the expert experience database involves: collecting full life-cycle data from multiple similar water plant photovoltaic and energy storage projects in China, and establishing a database linking technical solutions, environmental factors, and benefit data; similar water plant photovoltaic and energy storage projects need to cover different climate zones and water plant scales; the full life-cycle data includes technical solutions, environmental factors, economic benefits, and sustainability data; The technical solution includes photovoltaic and energy storage configuration parameters and installation process; environmental factors include local sunshine hours, temperature changes, and precipitation days; economic benefits include actual return on investment and payback period; sustainability data includes actual carbon emissions; typical deviation cases and corresponding solutions are marked in the associated database; typical deviation cases include insufficient power generation and overrun of operation and maintenance costs; training of the artificial intelligence analysis model: using the calculation data from steps S2-S6 as input features, and the deviation value between the actual operating data and the calculation data as the output label, the model is trained using an LSTM neural network; the calculation data includes load characteristics, photovoltaic and energy storage configuration, and economic parameters; the deviation values ​​include power generation deviation, cost deviation, and carbon emission deviation; During training, the training and test sets are divided in a 7:3 ratio. Cross-validation is used to adjust the number of network layers and hidden units to optimize the model's generalization ability. The number of network layers is 3 to 5, and the number of hidden units is 64 to 128. Deviation analysis and optimization suggestion generation: By comparing actual operating data with calculated data, the model outputs the top 3 core causes of deviation contribution rate; core causes include the fact that the number of rainy days in summer exceeds the calculated value, resulting in a 45% deviation contribution rate in photovoltaic power generation; solutions for similar deviation cases are matched from the expert experience database, and the details of the solution are adjusted in combination with the actual operating conditions of the current project to generate executable optimization suggestions; actual operating conditions include local climate and equipment status.

[0014] Preferably, in step S5, the estimation of the total project investment must follow these rules: The estimation accuracy of equipment and installation costs is controlled within ±5%; the unit price of photovoltaic modules is calculated based on the current market price, which includes transportation costs. The unit price of energy storage batteries is calculated based on capacity, and the price per unit capacity includes the BMS system. The installation fee includes labor and auxiliary material costs; In the construction cost, the photovoltaic support foundation is calculated as concrete usage × local concrete unit price + steel reinforcement usage × local steel reinforcement unit price + labor cost; The civil engineering cost of the energy storage room is calculated based on the building area multiplied by the local unit cost of similar industrial buildings, which includes decoration. Among other construction costs, land occupation fees will be charged on land owned by the water plant, with no fees charged for owned land; if additional land needs to be leased, the fee will be calculated as annual rent multiplied by the project calculation period. Survey and design fees are calculated according to the fee schedule; Supervision fees are calculated at 1.5% to 2.5% of the project cost, which is the sum of equipment and installation costs and construction costs. Production preparation costs include personnel training costs and spare parts costs; personnel training costs are calculated based on 3 to 5 people × training cycle × average training cost per person. Spare parts and spare parts costs are calculated at 1% to 2% of the equipment purchase cost; contingency costs are calculated at 5% to 8% of the sum of equipment and installation costs, construction costs, and other construction costs. Contingency costs are used to cope with risks such as price fluctuations and design changes during the project construction period. Total project investment = equipment and installation costs + construction costs + other construction costs + contingency fund.

[0015] This invention provides a smart assessment method and system for investment in waterworks photovoltaic and energy storage projects. It offers the following advantages: By leveraging Levy's global exploration capabilities, it avoids the limitations of traditional K-medoids algorithms that are prone to getting stuck in local optima. It can accurately extract multiple typical daily load curves from high-dimensional, noisy water load time series, effectively identifying hidden and dispersed load clustering patterns. It not only microscopically analyzes the matching relationship between each typical daily load and its corresponding illumination characteristics, but also macroscopically integrates the overall characteristics and variation patterns of the annual load. This ensures that the photovoltaic capacity configuration and energy storage unit selection can be deeply adapted to the actual energy needs of water plants, significantly improving the scientific nature of photovoltaic and energy storage system planning and the accuracy of economic prediction, thus avoiding configuration mismatch and investment deviations from the source. By systematically reviewing the full lifecycle data of completed photovoltaic and energy storage projects, a quantitative evaluation system is constructed based on technical, economic, and sustainability dimensions. By comparing and analyzing actual operational data with previous calculation results, and relying on an expert experience database and professionally trained artificial intelligence analysis models, deviation attribution and deep optimization strategies are generated to ensure the continuous and stable operation of photovoltaic and energy storage projects throughout their entire lifecycle and promote the continuous improvement of comprehensive benefits.

[0016] During the operation phase, it can achieve centralized management and control of energy efficiency monitoring, intelligent scheduling, fault diagnosis, and optimized operation and maintenance, completely changing the situation of various functions being isolated from each other in traditional solutions. It can significantly improve the management efficiency and decision-making intelligence level of water plant photovoltaic storage projects, and provide powerful tool support for energy conservation, cost reduction and efficiency improvement in the industry. Attached Figure Description

[0017] Figure 1 This is a flowchart of the evaluation system of the present invention; Figure 2 This is a flowchart of the load analysis process of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] like Figure 1 and Figure 2 This invention proposes an intelligent investment assessment system for water plant photovoltaic storage construction projects, comprising a hardware layer and a functional layer. The components of the hardware layer use industrial standard communication protocols to achieve data interaction and command transmission. It should be noted that the specific protocols include Modbus, OPCUA, or EtherCAT, which are commonly used protocols in water plant energy management scenarios, taking into account both compatibility and transmission efficiency. The hardware layer includes sensor networks, monitoring centers, integrated management platforms, algorithm servers, equipment control centers, and central databases. Sensor networks are deployed at water plant load nodes, photovoltaic module arrays, and energy storage battery packs to collect real-time power load data, photovoltaic output data, and energy storage system operating status data. The monitoring center communicates with the sensor network to perform noise reduction and missing value completion preprocessing on the collected data. The integrated management platform communicates bidirectionally with the monitoring center, algorithm server, equipment control center, and central database. It serves as the core hub for system data flow and control command issuance. The central database is used to store historical power load / photovoltaic output / energy storage status data collected by the sensor network, calculation and evaluation results output by the algorithm server, and expert experience database data. It also supports data retrieval from the integrated management platform and result backtracking from the algorithm server. The calculation and evaluation module is implemented by the integrated management platform and the algorithm server. It is used to perform load characteristic analysis, photovoltaic and storage system optimization configuration, project cost estimation, full life cycle economic evaluation and post-commissioning evaluation of water plant photovoltaic and storage projects. The intelligent management module is implemented independently by the integrated management platform and is used to realize the visualization of key project data, support operational decision-making, and control system access security.

[0020] As an optional embodiment: the calculation and evaluation module includes the following sub-modules: The basic information integration submodule is used to input and store system basic parameters, financial basic parameters, and operational basic parameters; The system's basic parameters include photovoltaic module efficiency, energy storage battery type, building installation area, and module size; Financial parameters include benchmark rate of return, loan interest rate, and tax rate; operational parameters include annual equipment degradation rate, annual maintenance cost, and energy storage battery cycle life. The load characteristic analysis submodule is used to extract representative daily load curves from the high-dimensional noisy power load time series data of the water plant using the K-medoids clustering algorithm based on Levy flight; the photovoltaic optimization configuration submodule is used to calculate the maximum installable capacity and recommended configuration capacity of photovoltaic modules by combining the priority of the building's installable area, local solar resource data and the water plant's load absorption capacity. The priority of areas where buildings can be installed is divided into the highest priority and the second highest priority. The energy storage optimization configuration submodule is used to determine the rated power, rated capacity, and charging and discharging strategy of the energy storage system based on the photovoltaic output curve, water plant load characteristics, peak-valley electricity pricing policy, and inverter total capacity constraints, and to perform physical layout feasibility verification; the cost estimation submodule is used to estimate the equipment purchase and installation costs, construction costs, and other construction costs according to the photovoltaic and energy storage system configuration results, and to form the total investment estimate of the project. The economic evaluation submodule is used to build dynamic financial models and calculate the project's internal rate of return, payback period, and total return on investment. The post-evaluation submodule is used to compare the actual operating data after the project is put into operation with the preliminary calculation results, and output targeted optimization suggestions.

[0021] As an optional embodiment: when the load characteristic analysis submodule executes the K-medoids clustering algorithm based on Levy flight improvement, it includes the following steps: Step 201: Randomly select K initial Medoids, where K ranges from 3 to 8, calculate the Manhattan distance from each load data point to the initial Medoids, and determine the initial clusters and the total clustering cost; Step 202: Assign all points to the nearest Medoids; The formula for calculating Manhattan distance is: ; Where xᵢ is the i-th load data point, m j Let t be the j-th Medoid, and t be the time dimension; For each data point xᵢ, calculate its path to all Medoids. The distance to the Medoid is calculated, and the current total cost is the sum of the distances between all sample points and the Medoid. The specific formula is as follows: ; Step 203: Iterate through all non-Medoid sample points, and try to replace each current Medoid with a non-Medoid sample point in turn, calculating the cost change after the replacement. ,in The total cost after replacement. Total cost before replacement; Step 204: If ΔCost < 0, perform the optimal replacement, which is to select the scheme with the smallest ΔCost; if ΔCost ≥ 0, generate a random number in the interval between 0 and 1. If the random number is less than the preset trigger probability p, randomly select 1 from the current Medoids as the perturbation reference point.

[0022] It should be noted that the preset trigger probability p is determined based on the noise intensity of the water plant load data and the number of clusters K: when the load data noise intensity is >15% (the noise intensity is calculated based on the standard deviation / mean of the data), p is 0.2-0.3; when the load data noise intensity is ≤15%, p is 0.1-0.2; at the same time, when the number of clusters K=3-5, p is 0.15-0.3, and when K=6-8, p is 0.1-0.25, ensuring that the algorithm can avoid local optima under different data quality and clustering requirements, and the overall value of p is in the range of 0.1 to 0.3.

[0023] As an optional embodiment: when the load characteristic analysis submodule executes the K-medoids clustering algorithm based on Levy flight improvement, determining whether to trigger Levy flight includes the following steps: Step 205: To avoid the algorithm getting stuck in a local optimum, set the Levy flight trigger probability p and generate a random number between (0,1). If the random number is less than p, it can be considered that the Levy flight has been triggered and proceed to step 206; otherwise, proceed to step 209. Step 206: Select the disturbance center; Randomly select a Medoid from the current Medoids set M. m p As a reference point for disturbance; Step 207: Levy flight generates new candidate points; First, based on the Levy distribution, random Levy numbers need to be generated, and these will be used as the step size for Levy flight: ; ; ; ; Where u and v are normally distributed random variables; β1 is the characteristic index of the Levy distribution; Let v be the standard deviation of v; SL be the Levy flight stride. It is a gamma function; Map the continuous step size to discrete data point indices, and calculate the distance from all points to m. p Sort the distances and construct the distance-sorted sequence R: ; Where n is the total number of sample points in the dataset; Then, the ranking position r based on the Levy step size is calculated using the following formula: ; The final new Medoid index is determined as follows: ; in New candidate points selected based on Levy flight; Step 208: Force update the Medoid collection; Use new candidate points replace m p And update the Medoids set M to complete this perturbation update; Step 209: If the Medoids set changes, return to step 202 and start the next iteration; otherwise, end the algorithm and output the final result, which includes K clustering results. Finally, the Medoids set M and the representative load curves for each cluster are obtained.

[0024] As an optional embodiment: In the photovoltaic optimization configuration submodule, the highest priority of the building installation area is the space above the pool, the idle open space in the factory, and the unobstructed concrete roof; the next priority is the partially obstructed concrete roof and the corrugated steel roof. The formula for calculating the maximum number of photovoltaic modules that can be installed is: ; in Let be the effective area of ​​the installation area for the i-th building. is the photovoltaic module layout coefficient for the i-th installation area; in this embodiment, the value is 0.85-0.95, determined based on the flatness and shading conditions of the area; L PV The length of the photovoltaic module, l PV The width of photovoltaic modules, Ag PV Photovoltaic module installation azimuth angle, β Photovoltaic module installation tilt angle α The solar altitude angle of the area where the water plant is located. w PV The width of the photovoltaic module; The theoretical maximum total installed capacity PPVmax is calculated using the following formula: ; in PV,p This refers to the power of a single photovoltaic module.

[0025] Preferred configuration: The photovoltaic module has an initial capacity of P PV,RC The final calculation formula is as follows: ; Take immediately and The smaller value in the range.

[0026] It should be noted that, in order to avoid investment waste and curtailment caused by blindly pursuing full-area installation, [the following measures will be taken]. Compared with the recommended configuration capacity determined based on load analysis To make a comparison, The principle for determining the value is to achieve the best match between the photovoltaic power generation curve and the plant's load curve under the typical daily curve. The economic critical point is usually to meet 85% to 95% of the daytime load demand or to ensure that the total annual power generation does not exceed 110% of the water plant's annual electricity consumption.

[0027] As an optional embodiment: In the energy storage optimization configuration submodule, the rated discharge power of the energy storage system meets the following constraints: ; in The rated discharge power of the energy storage system. P PV Recommended capacity configuration for photovoltaic systems. This refers to the total capacity of the inverter at the point of common coupling.

[0028] The rated capacity of the energy storage system is determined through a time-series simulation model. It should be noted that the input data for the time-series simulation model includes the typical daily load curve for the whole year obtained based on load characteristic analysis, the local hourly photovoltaic output prediction curve, and the time-of-use electricity price table. The time-series simulation model quantifies and evaluates the peak-shaving effect of photovoltaic self-consumption on peak-hour load by simulating the basic operating strategy of charging during off-peak hours and discharging during peak hours. The quantitative evaluation indicators for the peak-shaving effect include the peak load reduction rate (calculated as: (average peak load before optimization - average peak load after optimization) / average peak load before optimization × 100%) and the daily profit from peak-valley electricity price arbitrage (calculated as: (peak electricity price - off-peak electricity price) × daily effective charging and discharging capacity). Accurately calculate the effective capacity (EES) of the energy storage system to achieve 1 to 2 complete charge-discharge cycles per day within the remaining peak-valley price difference space; After completing the initial configuration of electrical parameters, a feasibility and safety check of the physical layout is required. The check includes confirming that the installation location has sufficient load-bearing area, meets the requirements of safe fire prevention distance, and facilitates line connection and operation and maintenance. If these requirements are not met, the power or capacity configuration of the energy storage system should be adjusted until both the electrical parameters and the physical layout plan are feasible.

[0029] It should be noted that, based on the completion of load characteristic analysis and optimized configuration of the photovoltaic system, the optimized configuration of the energy storage system is then carried out. Considering the specific characteristics of the water plant scenario, the energy storage configuration needs to go beyond traditional theoretical constraints and must also focus on inverter capacity synergy, the impact of photovoltaic power generation on the energy storage revenue model, and the feasibility of equipment placement. The specific implementation is as follows: First, it is necessary to determine the core operating strategy and optimization objectives of the energy storage system. Specifically, this means clarifying the basic operating strategy of the energy storage system under the main revenue model of peak-valley electricity price arbitrage, i.e., charging during off-peak hours and discharging during peak hours. The optimization objective is to maximize the net return of the energy storage system over its entire life cycle, or to achieve specific investment payback period requirements, while meeting technical constraints.

[0030] As an optional embodiment: the intelligent management module includes the following sub-modules: The project statistics dashboard module is used to visually display basic project information, investment cost details, key operational indicators and economic evaluation indicators through bar charts, line trend charts, data tables and ranking lists. It supports filtering data by daily, weekly, monthly and yearly time dimensions. Basic project information includes construction scale and construction period; investment cost details include equipment cost, construction cost, and other cost percentages; key operational indicators include photovoltaic self-consumption rate and energy storage charging and discharging efficiency; economic evaluation indicators include internal rate of return and investment payback period. The cockpit submodule is used to integrate an electronic map engine to mark the geographical coordinates of each water plant. The map displays the real-time power generation of each water plant by default using color gradients. The color gradient division standard is red > 80% of rated power, yellow 40% to 80% of rated power, and green < 40% of rated power. Clicking on a single water plant icon triggers a floating data card that displays the plant's real-time power generation, real-time power consumption, power surplus or deficit, photovoltaic self-consumption rate, and remaining energy storage capacity or charging / discharging progress. It also provides a quick access point to the detailed power generation and power consumption statistics analysis interface for that water plant. The user management submodule is used to input, edit, and query user information, as well as disable or enable accounts. It supports user password reset and records user login logs. User information includes username, department, contact number, and job title. User login logs include login time, login IP, and login status. The system management submodule is used to configure a multi-level tree-structured department architecture, define role permissions, assign roles to users, and record user operation logs. It supports searching operation logs by keywords. The multi-level tree-structured department architecture supports department merging and splitting operations. Role permissions are associated with system menus, operation buttons, and the viewing scope of monitoring point data. User operation logs include operation time, operator, operation content, and operation result.

[0031] As an optional embodiment, it also includes a smart assessment method for investment in waterworks photovoltaic-storage construction projects, comprising the following steps: S1: Input system basic parameters, financial basic parameters, and operational basic parameters through the basic information integration submodule and store them in the central database to provide data support for subsequent analysis. Among them, the system basic parameters include at least photovoltaic module efficiency, energy storage battery type, building installation area and module size; the financial basic parameters include at least benchmark rate of return, loan interest rate and tax rate; and the operational basic parameters include at least equipment annual degradation rate, annual operation and maintenance cost and energy storage battery cycle life. S2: The load characteristic analysis submodule calls the algorithm server to execute the K-medoids clustering algorithm based on Levy flight improvement. Input the water plant's historical power load data for the past year or more, and output K typical daily load curves and the annual frequency of each curve; the historical power load data must cover the energy consumption characteristics of the entire year. S3: Import building installation area survey data through the photovoltaic optimization configuration submodule, determine area priority, calculate the maximum installable capacity PPVmax and recommended configuration capacity PPVRC of photovoltaic modules, and generate a photovoltaic system configuration scheme; the building installation area survey data includes the area of ​​each area, shading status, and load-bearing capacity; the photovoltaic system configuration scheme includes module selection, installation area distribution, and array arrangement diagram. S4: Import the typical daily load curve output from step S2, the photovoltaic output prediction curve output from step S3, and the local time-of-use electricity price data through the energy storage optimization configuration submodule. Determine the preliminary electrical parameters of the energy storage system based on inverter capacity constraints and time-series simulation models. Combine the parameters with physical layout feasibility verification and adjustment to finally generate the energy storage system configuration scheme. The preliminary electrical parameters of the energy storage system include rated power PESdcmax and rated capacity EES. The energy storage system configuration scheme includes battery type, charging and discharging strategy, installation location planning, and fire protection design requirements. S5: Based on the configuration schemes in steps S3 and S4, the cost estimation submodule estimates the equipment purchase and installation costs, construction costs, and other construction costs by category, adds contingency funds to form the total project investment, and outputs the total investment estimate table. S6: Construct a dynamic financial model through the economic evaluation submodule. Input the total investment in step S5, the annual photovoltaic power generation revenue in step S3, the energy storage peak-valley arbitrage revenue in step S4, the annual operation and maintenance costs and tax rates in the basic operation parameters, calculate the annual cash flow according to the calculation period of conventional projects in the industry, and output the internal rate of return on investment, investment payback period, total investment return rate and net present value. S7: After the project is put into operation, the actual operation data is collected through the post-evaluation submodule and compared with the calculation results in step S6. The actual operation data includes monthly power generation, load factor, energy storage charging and discharging efficiency, operation and maintenance costs, and carbon emissions. Optimization suggestions include photovoltaic module tilt angle adjustment, energy storage charging and discharging time period optimization, and operation and maintenance cycle adjustment. S8: The project statistics dashboard of the intelligent management module and the cockpit sub-module display key data of steps S2-S7 in real time. Access control is achieved through user management and system management sub-modules to ensure the safe operation of the system and the compliant use of data.

[0032] It should be noted that the construction of the expert experience database involves: collecting full life-cycle data from multiple similar water plant photovoltaic and energy storage projects in China, and establishing a database linking technical solutions, environmental factors, and benefit data; similar water plant photovoltaic and energy storage projects need to cover different climate zones and water plant scales; the full life-cycle data includes technical solutions, environmental factors, economic benefits, and sustainability data; The technical solution includes photovoltaic and energy storage configuration parameters and installation process; environmental factors include local sunshine hours, temperature changes, and precipitation days; economic benefits include actual return on investment and payback period; sustainability data includes actual carbon emissions; typical deviation cases and corresponding solutions are marked in the associated database; typical deviation cases include insufficient power generation and overrun of operation and maintenance costs; training of the artificial intelligence analysis model: using the calculation data from steps S2-S6 as input features, and the deviation value between the actual operating data and the calculation data as the output label, the model is trained using an LSTM neural network; the calculation data includes load characteristics, photovoltaic and energy storage configuration, and economic parameters; the deviation values ​​include power generation deviation, cost deviation, and carbon emission deviation; During training, the training and test sets are divided in a 7:3 ratio. Cross-validation is used to adjust the number of network layers and hidden units to optimize the model's generalization ability. The number of network layers is 3 to 5, and the number of hidden units is 64 to 128. Deviation analysis and optimization suggestion generation: By comparing actual operating data with calculated data, the model outputs the top 3 core causes of deviation contribution rate; core causes include the fact that the number of rainy days in summer exceeds the calculated value, resulting in a 45% deviation contribution rate in photovoltaic power generation; solutions for similar deviation cases are matched from the expert experience database, and the details of the solution are adjusted in combination with the actual operating conditions of the current project to generate executable optimization suggestions; actual operating conditions include local climate and equipment status.

[0033] As an optional implementation: In step S5, the estimation of the total project investment must follow the following rules: The estimation accuracy of equipment and installation costs is controlled within ±5%; the unit price of photovoltaic modules is calculated based on the current market price, which includes transportation costs. The unit price of energy storage batteries is calculated based on capacity, and the price per unit capacity includes the BMS system. The installation fee includes labor and auxiliary material costs; In the construction cost, the photovoltaic support foundation is calculated as concrete usage × local concrete unit price + steel reinforcement usage × local steel reinforcement unit price + labor cost; The civil engineering cost of the energy storage room is calculated based on the building area multiplied by the local unit cost of similar industrial buildings, which includes decoration. Among other construction costs, land occupation fees will be charged on land owned by the water plant, with no fees charged for owned land; if additional land needs to be leased, the fee will be calculated as annual rent multiplied by the project calculation period. Survey and design fees are calculated according to the fee schedule; Supervision fees are calculated at 1.5% to 2.5% of the project cost, which is the sum of equipment and installation costs and construction costs. Production preparation costs include personnel training costs and spare parts costs; personnel training costs are calculated based on 3 to 5 people × training cycle × average training cost per person. Spare parts and spare parts costs are calculated at 1% to 2% of the equipment purchase cost; contingency costs are calculated at 5% to 8% of the sum of equipment and installation costs, construction costs, and other construction costs. Contingency costs are used to cope with risks such as price fluctuations and design changes during the project construction period. Total project investment = equipment and installation costs + construction costs + other construction costs + contingency fund.

[0034] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A smart investment assessment system for waterworks power plant photovoltaic storage projects, characterized in that, It includes a hardware layer and a functional layer. The components in the hardware layer use industry-standard communication protocols to achieve data interaction and command transmission. The hardware layer includes sensor networks, monitoring centers, integrated management platforms, algorithm servers, equipment control centers, and central databases. Sensor networks are deployed at water plant load nodes, photovoltaic module arrays, and energy storage battery packs to collect real-time power load data, photovoltaic output data, and energy storage system operating status data. The monitoring center communicates with the sensor network to perform noise reduction and missing value completion preprocessing on the collected data. The integrated management platform communicates bidirectionally with the monitoring center, algorithm server, equipment control center, and central database. It serves as the core hub for system data flow and control command issuance. The central database is used to store historical power load / photovoltaic output / energy storage status data collected by the sensor network, calculation and evaluation results output by the algorithm server, and expert experience database data. It also supports data retrieval from the integrated management platform and result backtracking from the algorithm server. The calculation and evaluation module is implemented by the integrated management platform and the algorithm server. It is used to perform load characteristic analysis, photovoltaic and storage system optimization configuration, project cost estimation, full life cycle economic evaluation and post-commissioning evaluation of water plant photovoltaic and storage projects. The intelligent management module is implemented independently by the integrated management platform and is used to realize the visualization of key project data, support operational decision-making, and control system access security.

2. The intelligent investment assessment system for waterworks and solar power storage projects according to claim 1, characterized in that: The calculation and evaluation module includes the following sub-modules: The basic information integration submodule is used to input and store system basic parameters, financial basic parameters, and operational basic parameters; The system's basic parameters include photovoltaic module efficiency, energy storage battery type, building installation area, and module size; Financial parameters include benchmark rate of return, loan interest rate, and tax rate; operational parameters include annual equipment degradation rate, annual maintenance cost, and energy storage battery cycle life. The load characteristic analysis submodule is used to extract representative daily load curves from the high-dimensional noisy power load time series data of the water plant using the K-medoids clustering algorithm based on Levy flight; the photovoltaic optimization configuration submodule is used to calculate the maximum installable capacity and recommended configuration capacity of photovoltaic modules by combining the priority of the building's installable area, local solar resource data and the water plant's load absorption capacity. The priority of areas where buildings can be installed is divided into the highest priority and the second highest priority. The energy storage optimization configuration submodule is used to determine the rated power, rated capacity, and charging and discharging strategy of the energy storage system based on the photovoltaic output curve, water plant load characteristics, peak-valley electricity pricing policy, and inverter total capacity constraints, and to perform physical layout feasibility verification; the cost estimation submodule is used to estimate the equipment purchase and installation costs, construction costs, and other construction costs according to the photovoltaic and energy storage system configuration results, and to form the total investment estimate of the project. The economic evaluation submodule is used to build dynamic financial models and calculate the project's internal rate of return, payback period, and total return on investment. The post-evaluation submodule is used to compare the actual operating data after the project is put into operation with the preliminary calculation results, and output targeted optimization suggestions.

3. The intelligent investment assessment system for waterworks power plant photovoltaic storage projects according to claim 2, characterized in that: When the load characteristic analysis submodule executes the K-medoids clustering algorithm based on the Levy flight improvement, it includes the following steps: Step 201: Randomly select K initial Medoids, where K ranges from 3 to 8, calculate the Manhattan distance from each load data point to the initial Medoids, and determine the initial clusters and the total clustering cost; Step 202: Assign all points to the nearest Medoids; The formula for calculating Manhattan distance is: ; Where xᵢ is the i-th load data point, m j Let t be the j-th Medoid, and t be the time dimension; For each data point xᵢ, calculate its path to all Medoids. The distance to the Medoid is calculated, and the current total cost is the sum of the distances between all sample points and the Medoid. The specific formula is as follows: ; Step 203: Iterate through all non-Medoid sample points, and try to replace each current Medoid with a non-Medoid sample point in turn, calculating the cost change after the replacement. ,in The total cost after replacement. Total cost before replacement; Step 204: If ΔCost < 0, perform the optimal replacement, which is to select the scheme with the smallest ΔCost; if ΔCost ≥ 0, generate a random number in the interval between 0 and 1. If the random number is less than the preset trigger probability p, randomly select 1 from the current Medoids as the perturbation reference point.

4. The intelligent investment assessment system for waterworks power plant photovoltaic storage projects according to claim 3, characterized in that: When the load characteristic analysis submodule executes the K-medoids clustering algorithm based on Levy flight improvement, the determination of whether Levy flight is triggered also includes the following steps: Step 205: To avoid the algorithm getting stuck in a local optimum, set the Levy flight trigger probability p and generate a random number between (0,1). If the random number is less than p, it can be considered that the Levy flight has been triggered and proceed to step 206; otherwise, proceed to step 209. Step 206: Select the disturbance center; Randomly select a Medoid from the current Medoids set M. m p As a reference point for disturbance; Step 207: Levy flight generates new candidate points; First, based on the Levy distribution, Levy random numbers need to be generated, and these will be used as the step size for Levy flight: ； ； ； ； Where u and v are normally distributed random variables; β1 is the characteristic index of the Levy distribution; Let v be the standard deviation of v; SL be the Levy flight stride. It is a gamma function; Map the continuous step size to discrete data point indices, and calculate the distance from all points to m. p Sort the distances and construct the distance-sorted sequence R: ； Where n is the total number of sample points in the dataset; Then, the ranking position r based on the Levy step size is calculated using the following formula: ； The final new Medoid index is determined as follows: ； in New candidate points selected based on Levy flight; Step 208: Force update the Medoid collection; Use new candidate points replace m p And update the Medoids set M to complete this perturbation update; Step 209: If the Medoids set changes, return to step 202 and start the next iteration; otherwise, end the algorithm and output the final result, which includes K clustering results. Finally, the Medoids set M and the representative load curves for each cluster are obtained.

5. The intelligent investment assessment system for waterworks and solar-storage projects according to claim 2, characterized in that: In the photovoltaic optimization configuration submodule, the highest priority for building installation areas are the space above the water tank, the idle open space in the factory, and the unobstructed concrete roof; the next highest priority are the partially obstructed concrete roof and the corrugated steel roof. The formula for calculating the maximum number of photovoltaic modules that can be installed is: ； in Let be the effective area of ​​the installation area for the i-th building. Let be the photovoltaic module layout coefficient for the i-th installation area; L PV The length of the photovoltaic module, l PV The width of photovoltaic modules, Ag PV Photovoltaic module installation azimuth angle, β Photovoltaic module installation tilt angle α The solar altitude angle of the area where the water plant is located. w PV The width of the photovoltaic module; The theoretical maximum total installed capacity PPVmax is calculated using the following formula: ； in PV,p This refers to the power of a single photovoltaic module.

6. The intelligent investment assessment system for waterworks power plant photovoltaic storage projects according to claim 5, characterized in that: The initial configuration capacity of the photovoltaic modules P PV,RC The final calculation formula is as follows: ； Take immediately and The smaller value in the range.

7. The intelligent investment assessment system for waterworks power plant photovoltaic-storage construction projects according to claim 2, characterized in that: In the energy storage optimization configuration submodule, the rated discharge power of the energy storage system meets the following constraints: ； in The rated discharge power of the energy storage system. P PV Recommended capacity configuration for photovoltaic systems. The total capacity of the inverter at the point of common coupling; The rated capacity of the energy storage system is determined through a time-series simulation model.

8. The intelligent investment assessment system for waterworks and solar-storage projects according to claim 1, characterized in that: The intelligent management module includes the following sub-modules: The project statistics dashboard module is used to visually display basic project information, investment cost details, key operational indicators and economic evaluation indicators through bar charts, line trend charts, data tables and ranking lists. It supports filtering data by daily, weekly, monthly and yearly time dimensions. Basic project information includes construction scale and construction period; investment cost details include equipment cost, construction cost, and other cost percentages; key operational indicators include photovoltaic self-consumption rate and energy storage charging and discharging efficiency; economic evaluation indicators include internal rate of return and investment payback period. The cockpit submodule is used to integrate an electronic map engine to mark the geographical coordinates of each water plant. The map displays the real-time power generation of each water plant by default using color gradients. Clicking on a single water plant icon triggers a floating data card that displays the plant's real-time power generation, real-time power consumption, power surplus or deficit, photovoltaic self-consumption rate, and remaining energy storage capacity or charging / discharging progress. It also provides a quick access point to the detailed power generation and power consumption statistics analysis interface for that water plant. The user management submodule is used to input, edit, and query user information, as well as disable or enable accounts. It supports user password reset and records user login logs. User information includes username, department, contact number, and job title. User login logs include login time, login IP, and login status. The system management submodule is used to configure a multi-level tree-structured department architecture, define role permissions, assign roles to users, and record user operation logs. It supports searching operation logs by keywords. The multi-level tree-structured department architecture supports department merging and splitting operations. Role permissions are associated with system menus, operation buttons, and the viewing scope of monitoring point data. User operation logs include operation time, operator, operation content, and operation result.

9. A smart evaluation method for investment in waterworks power plant photovoltaic-storage construction projects, applicable to the smart evaluation system for investment in waterworks power plant photovoltaic-storage construction projects as described in any one of claims 1 to 8, characterized in that: Includes the following steps: S1: Input system basic parameters, financial basic parameters, and operational basic parameters through the basic information integration submodule and store them in the central database to provide data support for subsequent analysis. Among them, the system basic parameters include at least photovoltaic module efficiency, energy storage battery type, building installation area and module size; the financial basic parameters include at least benchmark rate of return, loan interest rate and tax rate; and the operational basic parameters include at least equipment annual degradation rate, annual operation and maintenance cost and energy storage battery cycle life. S2: The load characteristic analysis submodule calls the algorithm server to execute the K-medoids clustering algorithm based on Levy flight improvement. Input the water plant's historical power load data for the past year or more, and output K typical daily load curves and the annual frequency of each curve; the historical power load data must cover the energy consumption characteristics of the entire year. S3: Import building installation area survey data through the photovoltaic optimization configuration submodule, determine area priority, calculate the maximum installable capacity PPVmax and recommended configuration capacity PPVRC of photovoltaic modules, and generate a photovoltaic system configuration scheme; the building installation area survey data includes the area of ​​each area, shading status, and load-bearing capacity; the photovoltaic system configuration scheme includes module selection, installation area distribution, and array arrangement diagram. S4: Import the typical daily load curve output from step S2, the photovoltaic output prediction curve output from step S3, and the local time-of-use electricity price data through the energy storage optimization configuration sub-module. Determine the preliminary electrical parameters of the energy storage system based on the inverter capacity constraints and the timing simulation model. Combine the parameters with the physical layout feasibility check and adjustment to finally generate the energy storage system configuration scheme. The preliminary electrical parameters of the energy storage system include the rated power PESdcmax and the rated capacity EES; the energy storage system configuration scheme includes battery type, charging and discharging strategy, installation location planning, and fire protection design requirements. S5: Based on the configuration schemes in steps S3 and S4, the cost estimation submodule estimates the equipment purchase and installation costs, construction costs, and other construction costs by category, adds contingency funds to form the total project investment, and outputs the total investment estimate table. S6: Construct a dynamic financial model through the economic evaluation submodule. Input the total investment in step S5, the annual photovoltaic power generation revenue in step S3, the energy storage peak-valley arbitrage revenue in step S4, the annual operation and maintenance costs and tax rates in the basic operation parameters, calculate the annual cash flow according to the calculation period of conventional projects in the industry, and output the internal rate of return on investment, investment payback period, total investment return rate and net present value. S7: After the project is put into operation, the actual operation data is collected through the post-evaluation submodule and compared with the calculation results in step S6. The actual operation data includes monthly power generation, load factor, energy storage charging and discharging efficiency, operation and maintenance costs, and carbon emissions. Optimization suggestions include photovoltaic module tilt angle adjustment, energy storage charging and discharging time period optimization, and operation and maintenance cycle adjustment. S8: The project statistics dashboard of the intelligent management module and the cockpit sub-module display key data of steps S2-S7 in real time. Access control is achieved through user management and system management sub-modules to ensure the safe operation of the system and the compliant use of data.

10. The intelligent investment assessment method for waterworks power plant photovoltaic storage projects according to claim 9, characterized in that, In step S5, the estimation of the total project investment must follow these rules: The estimation accuracy of equipment and installation costs is controlled within ±5%; the unit price of photovoltaic modules is calculated based on the current market price, which includes transportation costs. The unit price of energy storage batteries is calculated based on capacity, and the price per unit capacity includes the BMS system. The installation fee includes labor and auxiliary material costs; In the construction cost, the photovoltaic support foundation is calculated as concrete usage × local concrete unit price + steel reinforcement usage × local steel reinforcement unit price + labor cost; The civil engineering cost of the energy storage room is calculated based on the building area multiplied by the local unit cost of similar industrial buildings, which includes decoration. Among other construction costs, land occupation fees will be charged on land owned by the water plant, with no fees charged for owned land; if additional land needs to be leased, the fee will be calculated as annual rent multiplied by the project calculation period. Survey and design fees are calculated according to the fee schedule; Supervision fees are calculated at 1.5% to 2.5% of the project cost, which is the sum of equipment and installation costs and construction costs. Production preparation costs include personnel training costs and spare parts costs; personnel training costs are calculated based on 3 to 5 people × training cycle × average training cost per person. Spare parts and spare parts costs are calculated at 1% to 2% of the equipment purchase cost; contingency costs are calculated at 5% to 8% of the sum of equipment and installation costs, construction costs, and other construction costs. Contingency costs are used to cope with risks such as price fluctuations and design changes during the project construction period. Total project investment = equipment and installation costs + construction costs + other construction costs + contingency fund.

Images